[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

WO2014145929A1 - Systems and methods for addressing a media database using distance associative hashing - Google Patents

Systems and methods for addressing a media database using distance associative hashing Download PDF

Info

Publication number
WO2014145929A1
WO2014145929A1 PCT/US2014/030782 US2014030782W WO2014145929A1 WO 2014145929 A1 WO2014145929 A1 WO 2014145929A1 US 2014030782 W US2014030782 W US 2014030782W WO 2014145929 A1 WO2014145929 A1 WO 2014145929A1
Authority
WO
WIPO (PCT)
Prior art keywords
cue
value
patch
database
values
Prior art date
Application number
PCT/US2014/030782
Other languages
French (fr)
Inventor
Zeev Neumeier
Brian Reed
Original Assignee
Zeev Neumeier
Brian Reed
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US14/089,003 external-priority patent/US8898714B2/en
Priority to BR112015023389-9A priority Critical patent/BR112015023389B1/en
Priority to CA2906192A priority patent/CA2906192C/en
Priority to MX2015012510A priority patent/MX356884B/en
Priority to MX2015012512A priority patent/MX365827B/en
Priority to MX2020001441A priority patent/MX2020001441A/en
Priority to CN201480015936.XA priority patent/CN105052161B/en
Priority to PCT/US2014/030795 priority patent/WO2014145938A1/en
Priority to BR112015023380-5A priority patent/BR112015023380B1/en
Priority to CN201811395356.4A priority patent/CN110083739B/en
Priority to BR112015023369-4A priority patent/BR112015023369B1/en
Priority to PCT/US2014/030805 priority patent/WO2014145947A1/en
Priority to EP14762850.7A priority patent/EP3001871B1/en
Priority to MX2015012511A priority patent/MX366327B/en
Application filed by Zeev Neumeier, Brian Reed filed Critical Zeev Neumeier
Priority to CA2906199A priority patent/CA2906199C/en
Priority to CN201480017043.9A priority patent/CN105144141B/en
Priority to CA2906173A priority patent/CA2906173C/en
Priority claimed from PCT/US2014/030805 external-priority patent/WO2014145947A1/en
Priority claimed from US14/217,375 external-priority patent/US9094714B2/en
Priority claimed from US14/217,094 external-priority patent/US8930980B2/en
Priority claimed from PCT/US2014/030795 external-priority patent/WO2014145938A1/en
Priority to CA3173549A priority patent/CA3173549A1/en
Publication of WO2014145929A1 publication Critical patent/WO2014145929A1/en
Priority to MX2019007031A priority patent/MX2019007031A/en
Priority to CL2015002619A priority patent/CL2015002619A1/en
Priority to CL2015002623A priority patent/CL2015002623A1/en
Priority to MX2019008020A priority patent/MX2019008020A/en
Priority to HK16105168.7A priority patent/HK1218193A1/en
Priority to HK16105782.3A priority patent/HK1217794A1/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/4302Content synchronisation processes, e.g. decoder synchronisation
    • H04N21/4307Synchronising the rendering of multiple content streams or additional data on devices, e.g. synchronisation of audio on a mobile phone with the video output on the TV screen
    • H04N21/43074Synchronising the rendering of multiple content streams or additional data on devices, e.g. synchronisation of audio on a mobile phone with the video output on the TV screen of additional data with content streams on the same device, e.g. of EPG data or interactive icon with a TV program
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/783Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/7847Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using low-level visual features of the video content
    • G06F16/785Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using low-level visual features of the video content using colour or luminescence
    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11BINFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
    • G11B27/00Editing; Indexing; Addressing; Timing or synchronising; Monitoring; Measuring tape travel
    • G11B27/02Editing, e.g. varying the order of information signals recorded on, or reproduced from, record carriers
    • G11B27/031Electronic editing of digitised analogue information signals, e.g. audio or video signals
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04HBROADCAST COMMUNICATION
    • H04H60/00Arrangements for broadcast applications with a direct linking to broadcast information or broadcast space-time; Broadcast-related systems
    • H04H60/35Arrangements for identifying or recognising characteristics with a direct linkage to broadcast information or to broadcast space-time, e.g. for identifying broadcast stations or for identifying users
    • H04H60/37Arrangements for identifying or recognising characteristics with a direct linkage to broadcast information or to broadcast space-time, e.g. for identifying broadcast stations or for identifying users for identifying segments of broadcast information, e.g. scenes or extracting programme ID
    • H04H60/375Commercial
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04HBROADCAST COMMUNICATION
    • H04H60/00Arrangements for broadcast applications with a direct linking to broadcast information or broadcast space-time; Broadcast-related systems
    • H04H60/56Arrangements characterised by components specially adapted for monitoring, identification or recognition covered by groups H04H60/29-H04H60/54
    • H04H60/59Arrangements characterised by components specially adapted for monitoring, identification or recognition covered by groups H04H60/29-H04H60/54 of video
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/23Processing of content or additional data; Elementary server operations; Server middleware
    • H04N21/234Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs
    • H04N21/23418Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/25Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
    • H04N21/254Management at additional data server, e.g. shopping server, rights management server
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/25Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
    • H04N21/266Channel or content management, e.g. generation and management of keys and entitlement messages in a conditional access system, merging a VOD unicast channel into a multicast channel
    • H04N21/2668Creating a channel for a dedicated end-user group, e.g. insertion of targeted commercials based on end-user profiles
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/44Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
    • H04N21/44008Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/47End-user applications
    • H04N21/472End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content
    • H04N21/4722End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content for requesting additional data associated with the content
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/81Monomedia components thereof
    • H04N21/812Monomedia components thereof involving advertisement data
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/83Generation or processing of protective or descriptive data associated with content; Content structuring
    • H04N21/835Generation of protective data, e.g. certificates
    • H04N21/8358Generation of protective data, e.g. certificates involving watermark
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/83Generation or processing of protective or descriptive data associated with content; Content structuring
    • H04N21/845Structuring of content, e.g. decomposing content into time segments
    • H04N21/8456Structuring of content, e.g. decomposing content into time segments by decomposing the content in the time domain, e.g. in time segments
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/85Assembly of content; Generation of multimedia applications
    • H04N21/858Linking data to content, e.g. by linking an URL to a video object, by creating a hotspot

Definitions

  • This invention generally relates to the matching of unknown media data, such as video or audio segments, against a massive database of reference media files.
  • ACR automatic content recognition
  • index is often too large to reside in its entirety in the main memory of a computer server as used in a typical ACR system.
  • said database cannot fit entirely in the memory of a computer system, it is typically stored on magnetic disk storage and parts of said database are then read into memory in blocks corresponding to the index value providing the address. Said means of recalling partial database information is also known to one skilled in the art as "paging" which is a process common to many different computer software systems.
  • the present invention is an extension of the invention referenced above and is a system and method for matching unknown digital media such as television programing to a database of known media using a signal processing means employing a modified path pursuit algorithm, as described in the first invention.
  • Another novel aspect of the system and method as disclosed herein is its distance associative hash indexing means which can be subdivided into a plurality of independently addressable segments where each of said segments can address a portion of the database each of which can reside in its entirety in the main memory of a server means.
  • the resulting cluster of servers of the indexing means each hosts a sector of the index addressing associated data of a larger database of searchable audio or video information.
  • This indexing means of the invention results in a significant improvement in the speed and accuracy of the ACR system so enabled as to identify unknown media even when the television display is showing content where a user is changing channels, rewinding, fast-forwarding or even pausing video from a digital video recorder.
  • an exemplary method related to addressing a media database using distance associative hashing may include receiving one or more indications of a sample of a video segment; determining, for at least one patch of the sample of a video segment including at least one or more pixels of the at least one patch, an algorithmically-derived value of the one or more pixels of each patch; subtracting a median point value established for each patch from the mean value for each patch; transforming the values resulting from the subtraction using a function pre-derived to distribute the values evenly; constructing a hash value from the transformed values; referencing a number of most significant bits of the constructed hash value to determine a database sector; and storing at least the hash value on the determined database sector.
  • At least one of the receiving, determining, subtracting, transforming, constructing, referencing, or storing of the foregoing exemplary method is at least partially implemented using one or more processing devices.
  • receiving one or more indications of a sample of a video segment may include receiving one or more indications of at least one of a frame or a still image.
  • receiving one or more indications of a sample of a video segment may include receiving one or more indications of a sample of a video segment, the one or more indications of a sample of a video segment associated with at least one indication of a channel, at least one indication of a video segment, and at least one indication of a timecode offset from the beginning of the video segment.
  • determining, for at least one patch of the sample of a video segment including at least one or more pixels of the at least one patch, an algorithmically-derived value of the one or more pixels of each patch includes at least determining, for at least one patch of the sample of a video segment including at least one or more pixels of the at least one patch, a mean value of the one or more pixels of each patch.
  • subtracting a median point value established for each patch from the mean value for each patch may include subtracting a median point value established for each patch from the mean value for each patch, the median point value established for each patch having been previously determined utilizing data from each patch for a plurality of channels over at least one period of time.
  • transforming the values resulting from the subtraction using a function pre-derived to distribute the values evenly may include forming a variable matrix including at least the values resulting from the subtraction; obtaining a static matrix which, when crossed with the variable matrix, will more evenly distribute the transformed values; and computing a dot product of the variable matrix and the static matrix, the dot product including at least the more evenly-distributed transformed values.
  • obtaining a static matrix which, when crossed with the variable matrix, will more evenly distribute the transformed values may include determining, using locality-sensitive hashing at least partially based on one or more previously obtained hash values, a static matrix which, when crossed with a variable matrix, will more evenly distribute the transformed values of the variable matrix.
  • constructing a hash value from the transformed values may include constructing a hash value from the transformed values, including at least reducing the fidelity of the transformed values via reducing each transformed value to a binary representation.
  • reducing the fidelity of the transformed values via reducing each transformed value to a binary representation may include determining for each transformed value whether the transformed value is a positive number and, if the transformed value is a positive number, assigning a one to the hash value and otherwise assigning a zero to the hash value.
  • referencing a number of most significant bits of the constructed hash value to determine a database sector may include referencing a number of most significant bits of the constructed hash value to determine a database server, wherein the number of most significant bits is pre-determined to address a plurality of database servers, wherein a number of database servers associated with the number of most significant bits is established to enable at least one index associated with a database sector to reside entirely in memory of a corresponding database server.
  • storing at least the hash value on the determined database sector may include storing at least the hash value on the determined database sector, including at least storing at least one indication of a channel, at least one indication of a video segment, and at least one indication of a timecode offset from the beginning of the video segment at a database location at least partially based on the hash value.
  • circuitry and/or programming for effecting the herein-referenced method embodiments; the circuitry and/or programming can be virtually any combination of hardware, software, and/or firmware configured to effect the herein-referenced method aspects depending upon the design choices of the system designer.
  • an exemplary method related to addressing a media database using distance associative hashing may include receiving a cue, the cue constructed via one or more operations associated with a media storage operation; referencing a number of most significant bits of the received cue to determine a database sector; and returning at least one indication of at least one candidate from the database sector based at least partially on the received cue.
  • receiving a cue, the cue constructed via one or more operations associated with a media storage operation may include receiving a cue associated with a sample of a video buffer of a client system, including at least receiving one or more indications related to an epoch time associated with the sample of the video buffer of the client system.
  • receiving a cue, the cue constructed via one or more operations associated with a media storage operation may include receiving a cue, the cue associated with a sample of a video buffer of a client system, the cue at least partially determined by hashing at least some values associated with the video buffer.
  • receiving a cue, the cue associated with a sample of a video buffer of a client system, the cue at least partially determined by hashing at least some values associated with the video buffer may include receiving a cue, the cue associated with a sample of a video buffer of a client system, the cue at least partially determined by hashing at least some values associated with the video buffer, the hashing based at least partially one or more of at least one operand or at least one algorithm also utilized in an associated media storage operation.
  • receiving a cue, the cue constructed via one or more operations associated with a media storage operation may include receiving a cue, the cue determined via one or more operations including at least receiving one or more indications of at least one content of a video buffer of a client system; determining, for at least one patch of the at least one content of the video buffer including at least one or more pixels of the at least one patch, an algorithmically- derived value of the one or more pixels of each patch; subtracting a median point value from the mean value for each patch; transforming the values resulting from the subtraction; constructing a hash value from the transformed values; and associating the cue at least partially with the constructed hash value, wherein at least one of the determining, subtracting, transforming, or constructing operations utilize one or more of at least one operand or at least one algorithm also utilized in an associated media storage operation.
  • returning at least one indication of at least one candidate from the database sector based at least partially on the received cue may include returning at least one indication of at least one candidate from the database sector based at least partially on a probabilistic point location in equal balls ("PPLEB") algorithm as a function of the received cue.
  • returning at least one indication of at least one candidate from the database sector based at least partially on the received cue may include returning at least one indication of at least one candidate from the database sector based at least partially on the received cue, the at least one candidate being within a predetermined inverse percentage distribution radius of the received cue.
  • circuitry and/or programming for effecting the herein-referenced method embodiments; the circuitry and/or programming can be virtually any combination of hardware, software, and/or firmware configured to effect the herein-referenced method aspects depending upon the design choices of the system designer.
  • an exemplary method related to addressing a media database using distance associative hashing may include receiving at least one indication of at least one candidate and at least one indication of at least one cue; adding a token to a bin associated with at least one received candidate; and determining whether a number of tokens in a bin exceeds a value associated with a probability that a client system is displaying a particular video segment associated with at least one cue and, if the number of tokens in a bin exceeds a value associated with a probability that a client system is display a particular video segment associated with at least one cue, returning at least some data associated with the particular video segment based at least partially on the bin.
  • adding a token to a bin associated with at least one received candidate may include adding a token to a time bin associated with at least one received candidate.
  • adding a token to a bin associated with at least one received candidate may include determining a relative time, including at least subtracting a candidate time associated with the at least one candidate from an arbitrary time associated with the at least one cue; and adding a token to a time bin associated with the candidate based at least partially on the determined relative time.
  • the method may include removing one or more tokens from a time bin based at least partially on a time period elapsing.
  • circuitry and/or programming for effecting the herein-referenced method embodiments; the circuitry and/or programming can be virtually any combination of hardware, software, and/or firmware configured to effect the herein-referenced method aspects depending upon the design choices of the system designer.
  • an exemplary system related to addressing a media database using distance associative hashing may include, but is not limited to, one or more computing devices; and one or more instructions that, when executed on at least some of the one or more computing devices, cause at least some of the one or more computing devices to at least receive at least one stream of rasterized video; create at least one hash value associated with at least one sample of at least one received rasterized video stream; determine at least one database sector for storing a created at least one hash value; and store a created at least one hash value on at least one determined database sector.
  • an exemplary system related to addressing a media database using distance associative hashing may include, but is not limited to, one or more computing devices; and one or more instructions that, when executed on at least some of the one or more computing devices, cause at least some of the one or more computing devices to at least receive one or more indications associated with at least one video buffer of at least one client system; determine a cue based at least partially on the at least one video buffer and at least one epoch time associated with the at least one video buffer, wherein one or more of at least one operand or at least one function associated with determining the cue is also utilized in an associated media storage operation; reference a number of most significant bits of a determined cue to determine a database sector; and return at least one indication of at least one candidate from a determined database sector based at least partially on a determined cue.
  • an exemplary system related to addressing a media database using distance associative hashing may include, but is not limited to, one or more computing devices; and one or more instructions that, when executed on at least some of the one or more computing devices, cause at least some of the one or more computing devices to at least receive at least one indication of at least one candidate and at least one indication of at least one cue; add a token to a bin associated with at least one received candidate; and determine whether a number of tokens in a bin exceeds a value associated with a probability that a client system is receiving a particular video segment associated with at least one received cue and, if the number of tokens in a bin exceeds a value associated with a probability that a client system is receiving a particular video segment associated with at least one received cue, returning at least some data associated with the particular video segment based at least partially on the bin.
  • Figure 1 illustrates the construction of a sectored video matching database as taught by this invention which begins with initial video ingest or capture process which is then continuously updated.
  • a television display system 101 and its corresponding television display memory buffer 103 are shown for a potential embodiment of the system.
  • the allocation of pixel patches 102 and a calculation of a value 105, using certain algorithmic means known to those skilled in the art, is made for each pixel patch and a resulting data structure is created and then time-stamped make a "cue" 106 which may also have additional metadata associated with it.
  • Figure 2 illustrates the processing of the cue data 201 and the generation of the hash index 202 using the distance associative hashing process, further illustrating the sectored addressing scheme 203 to store data in related groups (buckets) 206.
  • Figure 3 illustrates the real-time capture of unknown television content for recognition from a connected television monitor or the like 301.
  • a pixel patch is defined as typically a square pixel area of the video buffer 303 with dimensions of perhaps ten pixels by ten rows of pixels 304, however, any reasonable shape and dimension may be used.
  • the number of pixel patch positions can be any number between ten and fifty locations within said video buffer and is processed 305 to send cue data 306 to the central server means.
  • Figure 4 illustrates the extraction of candidate cue values 401 from the reference (matching) database bucket 404 and supplying said cue values 403 to the path pursuit content matching process 402 as taught in the first invention referenced above.
  • Figure 5 illustrates the data structure of bins which hold tokens for scoring candidate values from the matching database. Said bins are "leaky” and tokens expire over time as the search process progresses through time.
  • Figure 6 illustrates a typical memory paging scheme as taught by prior art for accessing large databases.
  • Figure 7 illustrates the creation of a hash value involving several steps beginning with computing the median value of each of the multiplicity of points which make up the samples from a frame of video.
  • Figure 8 illustrates how the hash value is computed.
  • Figure 9 illustrates the beneficial results of using the median values of a pixel location as part of the process of computing the hash values.
  • Figure 9a illustrates the problem of not using a media value when partitioning a multi-dimensional dataset.
  • Figure 9b illustrates the benefit of finding a media value of a dataset.
  • Figure 10 illustrates an operational flow representing example operations related to addressing a media database using distance associative hashing.
  • Figure 11 illustrates an alternative embodiment of the operational flow of Figure 10.
  • Figure 12 illustrates an alternative embodiment of the operational flow of Figure 10.
  • Figure 13 illustrates an alternative embodiment of the operational flow of Figure 10.
  • Figure 14 illustrates an alternative embodiment of the operational flow of Figure 10.
  • Figure 15 illustrates an alternative embodiment of the operational flow of Figure 10.
  • Figure 16 illustrates an alternative embodiment of the operational flow of Figure 10.
  • Figure 17 illustrates an alternative embodiment of the operational flow of Figure 10.
  • Figure 18 illustrates an alternative embodiment of the operational flow of Figure 10.
  • Figure 19 illustrates an alternative embodiment of the operational flow of Figure 10.
  • Figure 20 illustrates a different operational flow representing example operations related to addressing a media database using distance associative hashing.
  • Figure 21 illustrates an alternative embodiment of the operational flow of Figure 20.
  • Figure 22 illustrates an alternative embodiment of the operational flow of Figure 20.
  • Figure 23 illustrates an alternative embodiment of the operational flow of Figure 20.
  • Figure 24 illustrates another operational flow representing example operations related to addressing a media database using distance associative hashing.
  • Figure 25 illustrates an alternative embodiment of the operational flow of Figure 24.
  • Figure 26 illustrates an alternative embodiment of the operational flow of Figure 24.
  • Figure 27 illustrates a system related to addressing a media database using distance associative hashing.
  • Figure 28 illustrates another system related to addressing a media database using distance associative hashing.
  • Figure 29 illustrates yet another system related to addressing a media database using distance associative hashing.
  • the first invention which relates to this invention is a system and method of matching unknown video to a database of known video using a novel signal processing means employing a modified path pursuit algorithm, among other means, as described in the aforementioned publication.
  • a novel means of the new invention is its Distance Associated Hashing with its attendant provision of utilizing a sectored-index database access.
  • Said indexing means provides a highly computationally-efficient means for matching an unknown media segment to a reference database of known media, such as audio or video content.
  • This indexing means of the invention results in a significant improvement in the speed and accuracy of the ACR system so enabled as to track the identity of media even when the television display is showing content where a user is changing channels, rewinding, fast- forwarding or even pausing video from a digital video recorder.
  • Both the building, updating, and the subsequent accessing of the media matching database will describe a system capable of generating and addressing a sectored database such that the database sectors can each reside in the main memory of a respective multiplicity of server means without resorting to a paging means within each of the respective server means.
  • This collective means of addressing a sectored database through locality sensitive hashing provides a significant improvement in efficiency of operation.
  • the construction of a sectored video matching database begins with the process as illustrated in Figure 1.
  • a television system 101 decodes a television signal and places the contents of each frame of video into a video frame buffer in preparation for the display or further processing of pixel information of the frame of video.
  • Said television system can be any television decoding system that can decode a television signal whether from a baseband or modulated television source and fill a video frame buffer with the decoded RGB values at the respective frame size as specified by the video signal.
  • Such systems are well known to one skilled in the art.
  • the system of the invention first builds and then continuously updates a reference database of television programming fingerprints described in the original application as cues or cue values.
  • the invention performs the acquisition of one or more patches of video 102 which are read from the video frame buffer 103.
  • Said video patches can be any arbitrary shape or pattern but for the purposes of this example shall be 10 pixels horizontally by 10 pixels vertically. Also for the sake of this example, assume that there are 25 pixel patch positions within the video frame buffer that are evenly distributed within the boundaries of said buffer, though they do not have to be evenly distributed.
  • Each pixel shall consist of a red, a green and a blue value, 104, typically represented by an eight bit binary value for each color for a total of 24 bits or three bytes per patch location.
  • This composite data structure is populated with the average pixel values from a number of pixel patch positions from the video buffer.
  • a pixel patch is defined as a typically square pixel area of the video buffer with dimensions of perhaps ten pixels by ten rows of pixels 304. The number of pixel patch positions might typically be between ten and fifty locations within the video buffer.
  • Epoch time is defined as the time in fractions of a second that have elapsed since midnight, January 1,1970 which is an accepted convention in computing systems, particularly with Unix (or Linux)-based systems.
  • Metadata may be included and together a data structure 106 is defined called a tagged fingerprint, "cue”, or a "point”, as taught in the original patent application.
  • metadata attributes might be derived from closed-captioning data from the currently displaying video program or it could be keywords extracted by means of a speech recognition system operating within the processor means of the television system which converts audio from the respective television program into text information. Said textual information may then be searched for relevant key words or sent in its entirety as part of the cue data structure to the central server means for further processing.
  • the cue records 201 are passed in Figure 2 to a hash function 202 that generates a hash value 203 using a locality sensitive hashing algorithm based on Probabilistic Point Location in Equal Balls algorithm (PPLEB).
  • PPLEB Probabilistic Point Location in Equal Balls algorithm
  • the ten by ten pixel patch 302 shown in this particular example would have one hundred pixels and is mathematically averaged resulting in a mean pixel value 305 for red, green and blue values, respectively.
  • any averaging function can be used in place of a simple mean.
  • a plurality of such pixel patches are extracted from the video frame. If, by way of example, 25 such pixel patches are extracted from the video frame, then the result will be a point representing a position in a 75-dimension space. The skilled person will know that such a large search space could require extensive computing resources to later locate, even approximately, said value in combination with the other 74 values representing one frame of video.
  • the creation of the hash value involves several steps beginning with computing an algorithmically-derived value of each point as shown in Figure 7, 701 to 775.
  • One useful means of algorithmically deriving said median value is found by summing each point of every frame of every program stream or channel of video maintained by the matching database over perhaps a 24 hour period. The median of each point is found from the summation process.
  • the next step in deriving the final hash value is to subtract the mean value from the point value of each respective point, row 801 minus row 802 equals row 803.
  • the result is a plus or minus values to which a pre-derived hashing function is applied.
  • the result of the point values minus the mean values of the respective points are arranged in a matrix to which a dot- product is calculated using a similar matrix constituting the pre-derived hash value (or key).
  • the result of the dot product of the two matrices is then further transformed to a one or zero value based on the sign of the product matrix element.
  • the skilled person would set positive values to one and negative values to zero.
  • the resulting hash value points to more or less evenly distributed values across the data storage area.
  • the hash value 203 can be further divided, Figure 2, such that the 'n' most significant bits 205 addresses one of the 2 n (2 A n) sectors of the database.
  • the remaining bits 206 address individual 'buckets' of the addressed sector of the database, which will be described in more detail later.
  • Figure 6 illustrates a typical paging approach.
  • Figure 6 assume the example system is attempting to match unknown data to a database of known data.
  • An index 602 is used to address only the portion of the data 605 that can fit in the main memory of the CPU 606. This data is searched and, if results are negative, then another segment of data is fetched into main memory 603 and searching continues.
  • Google search engine A notable example might be the considerable Google search engine.
  • the skilled person knows this system to be one of the largest computer systems built to date. The speed and accuracy of the Google search process is remarkable.
  • the Google search means is considerably different and not at all applicable to matching unknown media to a database of known media even if the two databases were the same very large size. This is because the Google search means employs the map-reduce algorithm which is designed for searching large databases of essentially unassociated data. While an advance over paging system, map-reduce is a computationally-intense process which also requires significant data communications bandwidth between the participating computer systems. In contrast, this invention is efficient in the use of processing and communications resources.
  • the distance associative hashing function provides a means to address a database in sectors such that the data of said addressing means fits in the main memory of an individual server means of group of servers.
  • Said grouping is accomplished by grouping the data related by distance in a multi-dimensional array into the same sector using the distance associative hashing step as a means to achieve said grouping.
  • the sector identification for addressing a data element is calculated from the hash index generated from said process by extracting a subset of the total bits of said hash function and using said subset to address the desired sector in which to store data in the reference database.
  • the hash index subset is the address of the sector that contains the distance associated hash values, called buckets in the first invention.
  • the remainder of the hash address is then used to address a bucket of the sector for storing the new data.
  • the sector address can be found by means of re-hashing the first hash value.
  • the distance associative hashing provides a means to address a very complex (multi-dimensional) database quickly by finding data that is not an exact match but rather is within a predetermined radius (distance associative) of the value sought. Importantly, sometimes this addressing means will result in no match at all. Where a business-oriented database cannot tolerate inaccuracy, a media matching system can readily tolerate missed matches and will simply continue the matching process upon the arrival of the next data received and taught in the first patent.
  • Data arrival from the unknown source that is to be determined by the ACR system is periodic, of course, but can be commanded by the system of the invention to arrive at differing intervals based on the requirement for accuracy or by requirements imposed by the state of the system such as when the system might be nearing overload and the sending clients are then commanded to send a lower sample rate.
  • a typical data reception rate might be 1/10 second intervals, for example.
  • the group of pixel values are derived from every frame of video from every video source that is to be part of the reference database.
  • the group of pixel values and are then appended with the broadcast time of the video program as well as with certain metadata, which is information about the program such as the content identification (ID), title of the program, actors name, time of airing, short synopsis, etc.
  • Said metadata is generally acquired from commercial electronic program guide sources.
  • Said array of processed pixel values with the addition of the timecode plus the metadata are then stored in the reference database and the address of said stored data is then added to the hash index at the respective hash value and sector ID value.
  • a second database index is built and maintained by using the content ID from the metadata as another means for addressing the reference database.
  • the process of building and continuously updating the database is continuous and the number of days of data maintained by the database is based on the needs of the user but for example might range from one day to one month.
  • the process of identifying an unknown video segment from data received from a multiplicity of client devices begins with a procedure similar to that used above for building the reference database.
  • this procedure involves a television monitor 301, such as a popular flat-screen HDTV typically of the type known as the smart TV wherein the TV contains a processing means with memory capable of executing application programs similar to the type found on common smartphones of today.
  • the system of the invention samples regions 302 of a video frame buffer 301 in typically a multiplicity of places. Said samples are of an identical size, shape and position to the pixel patches used in the process of building the reference database.
  • Each of the collected pixel patches is then algorithmically processed to produce a computed value for the red, green and blue values of each patch in a manner identical to the method used to create the reference database.
  • Said system of the invention calculates a distance associative hash index of the collected mean values identical to the content ingest function described above.
  • the resulting sector identification (ID) value is extracting as a subset of the total bits of the hash index also identically to the ingest system described above.
  • the remainder of the hash index is used to address the desired sector in which to search for all candidate (potential) matches belonging to the same bucket as the unknown data point.
  • the system of the invention will also collect candidates from the database responsible for said sector belonging to the potential content ID, using the content ID index, created during the ingest process as described above, to address reference cues around time radius r' of the timestamp (of the successfully matched candidate). Duplicate candidates are next removed as well as candidates that are too far from the unknown point by radius r, as taught in the first patent.
  • Each matching candidate 501 is assigned a data structure 502 in the memory of the matching system of the invention.
  • the data structure consists of, among other things, arbitrary time bins grouped by some arbitrary amount (e.g. approximately one second). For the sake of example, assume said data structure consists of one hundred bins representing ten seconds of video cue points. The bins are generally not equally spaced in time.
  • a relative time is calculated by subtracting candidate time from the arbitrary time of the unknown video.
  • Candidate time is the time of broadcast of each video cue associated with the candidate during the reference program airing.
  • Epoch time is well known to the skilled person and is typically employed in computer systems. Said time is calculated as the current number of units of time since January 1 of 1970.
  • the relative time of the actually matching candidates should be close to that value. Likewise, candidates that are not a good match are not likely to have relative times close to the 100 seconds of this example.
  • the system of the invention adds a token to the respective bin of the candidate data structure. Said system then repeats the process for the next candidate as described in the previous paragraph.
  • Another, and important, step for the scoring of results is to apply time discounting to all bins.
  • This is a relatively simple process that decrements the value in all bins by a small amount for each cycle of time.
  • the skilled person would recognize this as a "leaky bucket" method of scoring.
  • bins that are no longer being filled by means of matching cue points will ultimately decrement to zero over a number of cycles of said process.
  • bins that are filled slowly by random noise in the system will likewise be decremented.
  • time discounting ultimate clears bins that are filled by false-positive matches and random noise.
  • the skilled person would also clearly see that without said time discount binning, all bins would eventually fill to capacity and no results could be obtained from the process.
  • Said time discounting also decrements to zero any bins with levels, such as 503, that are above the matching threshold 504 when the video stream from the client television monitor is in any way changed by any of the following: changing channels, rewind, fast forward, pausing video, etc.
  • Figure 8 illustrates how the hash value is computed. First a median value of each pixel location contributing to the video fingerprint is found by summing the values of said location over a period of many days of collection values at said location from a plurality of television channels representative of the typical television programming to be identified by the invention. Once the median value is determined is can be used indefinitely as a constant without further calculation or adjustment.
  • the pixel value sent from the client to the server matching system is first processed by subtracting the median value of said pixel location.
  • the resulting value is stored with the other pixels locations of a video frame in matrix and an appropriate hashing function is applied to said matrix. Hash values are then derived from the resulting dot product.
  • Figure 9 illustrates the beneficial results of using the median values of a pixel location as part of the process of computing the hash values.
  • Chart 901 shows the resulting curve of the output of a typical un-optimized hash function with a relatively small number of hash values occupying a relatively narrow range on the left edge of the curve.
  • the resulting median value 902 is relatively low.
  • Chart 903 shows the favorable redistribution of hash values as a resulting of computing the median of each pixel location that participates in the matching process and applying said median value as part of the hashing function.
  • the distribution of hash values is more spread out with an attendant rise in the median value of all hash keys 904.
  • Figure 9a illustrates what happens to a dataset when a median value is not found prior to partitioning said dataset. If the system sampled sixteen pixel locations of each video frame and if each pixel location had a red, green and blue pixel value, there would be 64 dimensions (or axis) to the graph. For the sake of illustration, in this example, the dataset includes just two pixel sample points of a single video frame 906 and 908. Further, the example assumes just a single luminance value is obtained at each pixel point.
  • Figure 9b illustrates the benefit of finding the median value of each pixel location.
  • This example continues to use the assumption that the pixel values are a single luminance value from zero to 255, although absolute value is of no consequence to this method.
  • This example illustrates a simplistic assumption of the median value is 128 for both pixel locations.
  • Diagonal slice 909 moves to 909'. It is clear from the illustration that now all eight sectors contain data.
  • the data can be spliced more than once around each median point of the 48 dimension graph as required to partition said data such that said dataset resulting from said slice can be made to fit within the operational constraints of an individual computer server of the system. In any case, data will be found most of the time on the clockwise and counterclockwise side of each partition slice.
  • Figure 10 illustrates an operational flow 1000 representing example operations related to addressing a media database using distance associative hashing.
  • discussion and explanation may be provided with respect to the above-described examples of Figures 1 through 9, and/or with respect to other examples and contexts.
  • the operational flows may be executed in a number of other environments and contexts, and/or in modified versions of Figures 1 through 9.
  • the various operational flows are presented in the sequence(s) illustrated, it should be understood that the various operations may be performed in other orders than those which are illustrated, or may be performed concurrently.
  • Operation 1002 depicts receiving one or more indications of a sample of a video segment.
  • the indications may be associated with one or more pixel patches from an ingest system.
  • operation 1004 depicts determining, for at least one patch of the sample of a video segment including at least one or more pixels of the at least one patch, an algorithmically-derived value of the one or more pixels of each patch. For example, as shown in and/or described with respect to Figures 1 through 9, a mean value of the red pixels in each patch, the green pixels in each patch, and the blue pixels in each patch may be computed.
  • operation 1006 depicts subtracting a median point value established for each patch from the mean value for each patch.
  • a median value of each pixel location contributing to the video fingerprint may be found by summing the values of said location over a period of many days of collection values at said location from a plurality of television channels.
  • operation 1008 depicts transforming the values resulting from the subtraction using a function pre-derived to distribute the values evenly.
  • the values resulting from the subtraction populate a matrix.
  • a dot product of that matrix and a pre-derived static matrix may be computed.
  • the pre-derived static matrix may be determined prior to operational flow 1000 being instantiated, and may be optimized mathematically based on past ingested data such that matrices crossed with it will produce more evenly distributed results than results coming directly from the subtraction operation.
  • operation 1010 depicts constructing a hash value from the transformed values.
  • a hash value may be a string of bits.
  • operation 1012 depicts referencing a number of most significant bits of the constructed hash value to determine a database sector.
  • a number of bits may be predetermined so that the predetermined number of bits of a hash value are used for addressing one or more database sectors.
  • operation 1014 depicts storing at least the hash value on the determined database sector.
  • the hash value may be stored in a bucket, the bucket including other hash values which are mathematically near, where the hash values are associated at least with particular video segments and offsets.
  • Figure 11 illustrates alternative embodiments of the example operational flow 1000 of Figure 10.
  • Figure 11 illustrates an example embodiment where operational flow 1000 may include at least one additional operation. Additional operations may include operation 1102.
  • Operation 1102 illustrates at least one of the receiving 1002, determining 1004, subtracting 1006, transforming 1008, constructing 1010, referencing 1012, or storing 1014 operations being at least partially implemented using one or more processing devices.
  • one of the foregoing operations may be at least partially implemented using one or more computer processors.
  • Other processing devices may include Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs), digital signal processors (DSPs), or any other circuitry configured to effect the result of at least one of the foregoing operations.
  • ASICs Application Specific Integrated Circuits
  • FPGAs Field Programmable Gate Arrays
  • DSPs digital signal processors
  • Figure 12 illustrates alternative embodiments of the example operational flow 1000 of Figure 10.
  • Figure 12 illustrates an example embodiment where operation 1002 may include at least one additional operation. Additional operations may include operation 1202, and/or operation 1204.
  • Operation 1202 illustrates receiving one or more indications of at least one of a frame or a still image.
  • a sample of a video segment may be comprised of an individual frame of a video stream. Such a frame may be one 30fps video frame.
  • a sample of a video segment may be a still image, or a portion of a video segment that may be imaged at a rate other than 30 times a second.
  • operation 1204 illustrates receiving one or more indications of a sample of a video segment, the one or more indications of a sample of a video segment associated with at least one indication of a channel, at least one indication of a video segment, and at least one indication of a timecode offset from the beginning of the video segment.
  • data associated with a video segment (which may be a program title and/or other metadata associated with a video segment), the channel from which the program was ingested, and an offset in time from the start of the program may be received, from, for example, a channel guide associated with a channel which is being monitored by the ingest system.
  • Figure 13 illustrates alternative embodiments of the example operational flow 1000 of Figure 10.
  • Figure 13 illustrates an example embodiment where operation 1004 may include at least one additional operation 1302.
  • Operation 1302 illustrates determining, for at least one patch of the sample of a video segment including at least one or more pixels of the at least one patch, a mean value of the one or more pixels of each patch.
  • the algorithmic operation used to reduce the one or more pixels in a patch to a single value may be, for example, an arithmetic mean.
  • Figure 14 illustrates alternative embodiments of the example operational flow 1000 of Figure 10.
  • Figure 14 illustrates an example embodiment where operation 1006 may include at least one additional operation 1402.
  • Operation 1402 illustrates subtracting a median point value established for each patch from the mean value for each patch, the median point value established for each patch having been previously determined utilizing data from each patch for a plurality of channels over at least one period of time.
  • a median value may be determined, the median value determined for each patch, wherein medians are established for the same patches at ingest as in the operation of determining a segment on a client system, the median being established as a constant value derived from monitoring the same patches across many channels for a long time (a month, a year, etc.).
  • Figure 15 illustrates alternative embodiments of the example operational flow 1000 of Figure 10.
  • Figure 15 illustrates an example embodiment where operation 1008 may include at least one additional operation. Additional operations may include operation 1502, operation 1504, and/or operation 1506.
  • Operation 1502 illustrates forming a variable matrix including at least the values resulting from the subtraction. For example, as shown in and/or described with respect to Figures 1 through 9, values are arranged in a matrix, the values resulting from the subtraction operation, wherein the subtraction operation subtracts the median value established over time for each patch from the mean value of the instant frame being ingested.
  • Operation 1504 illustrates obtaining a static matrix which, when crossed with the variable matrix, will more evenly distribute the transformed values.
  • a matrix may be determined based upon mathematical analysis of previously-obtained data sets related to hash values.
  • the matrix may be optimized mathematically such that, when used as an operand in a dot product operation with successive variable matrices, the corresponding successive result matrices will include values that are more evenly spread along a distribution curve than the variable matrices prior to the dot product operation.
  • Operation 1506 illustrates computing a dot product of the variable matrix and the static matrix, the dot product including at least the more evenly-distributed transformed values.
  • the variable matrix containing values resulting from the subtraction operation may be crossed with a static matrix that has been predetermined to distribute data represented by a variable matrix more evenly, such that the resulting matrices are more spread out instead of being bunched about a particular portion of the distribution.
  • Figure 16 illustrates alternative embodiments of the example operational flow 1000 of Figure 10.
  • Figure 16 illustrates an example embodiment where operation 1504 may include at least one additional operation 1602.
  • Operation 1602 illustrates determining, using locality-sensitive hashing at least partially based on one or more previously obtained hash values, a static matrix which, when crossed with a variable matrix, will more evenly distribute the transformed values of the variable matrix.
  • a locality-sensitive hashing technique may be used to analyze previously-ingested video samples, producing a matrix such that, when used as an operand in a dot product operation with successive variable matrices, the corresponding successive result matrices will include values that are more evenly spread along a distribution curve than the variable matrices prior to the dot product operation.
  • Figure 17 illustrates alternative embodiments of the example operational flow 1000 of Figure 10.
  • Figure 17 illustrates an example embodiment where operation 1010 may include at least one additional operation. Additional operations may include operation 1702, and/or operation 1704.
  • Operation 1702 illustrates constructing a hash value from the transformed values, including at least reducing the fidelity of the transformed values via reducing each transformed value to a binary representation.
  • each value of the resultant matrix from the dot product operation may be reduced from, for example, an 8-bit value from 0 to 255 (or from -127 to 128) to a single bit, being either a one or a zero.
  • Operation 1702 may include operation 1704.
  • Operation 1704 illustrates determining for each transformed value whether the transformed value is a positive number and, if the transformed value is a positive number, assigning a one to the hash value and otherwise assigning a zero to the hash value. For example, as shown in and/or described with respect to Figures 1 through 9, each value of the resultant matrix from the dot product operation between 1 and 128 may be reduced to a bit value of 1, and each value of the resultant matrix from the dot product operation between -127 and 0 may be reduced to a bit value of 0.
  • Figure 18 illustrates alternative embodiments of the example operational flow 1000 of Figure 10.
  • Figure 18 illustrates an example embodiment where operation 1012 may include at least one additional operation 1802.
  • Operation 1802 illustrates referencing a number of most significant bits of the constructed hash value to determine a database server, wherein the number of most significant bits is pre-determined to address a plurality of database servers, wherein a number of database servers associated with the number of most significant bits is established to enable at least one index associated with a database sector to reside entirely in memory of a corresponding database server.
  • a number of most significant bits of 2 may be selected, whereby the 2 bits may provide four different values (00, 01, 10, and 11), each of which may be assigned to a different database sector.
  • the number of most significant bits of a hash value may be established to provide a sufficient number of servers such that a content associated with a plurality of hash values may fit entirely in the memory of a particular database sector, which may be a database server, a cluster partner, a virtual machine, and/or another type of database node.
  • the number of bits does not have to, but may, exactly represent the maximum number of database sectors at any given time (i.e. while 6 bits may be selected to provide for addressing of up to 64 database sectors, the system may be operable with fewer servers e.g. 60 sectors, or with the maximum 64 sectors).
  • Figure 19 illustrates alternative embodiments of the example operational flow 1000 of Figure 10.
  • Figure 19 illustrates an example embodiment where operation 1014 may include at least one additional operation 1902.
  • Operation 1902 illustrates storing at least the hash value on the determined database sector, including at least storing at least one indication of a channel, at least one indication of a video segment, and at least one indication of a timecode offset from the beginning of the video segment at a database location at least partially based on the hash value.
  • data associated with a video segment (which may be a program title and/or other metadata associated with a video segment), the channel from which the program was ingested, and an offset in time from the start of the program may be stored, either along with the hash value or in a location associated with and/or referenceable by the hash value, the storage being in the same or different sector, server, or database as the hash value.
  • Figure 20 illustrates an operational flow 2000 representing example operations related to addressing a media database using distance associative hashing.
  • discussion and explanation may be provided with respect to the above-described examples of Figures 1 through 9, and/or with respect to other examples and contexts.
  • the operational flows may be executed in a number of other environments and contexts, and/or in modified versions of Figures 1 through 9.
  • the various operational flows are presented in the sequence(s) illustrated, it should be understood that the various operations may be performed in other orders than those which are illustrated, or may be performed concurrently.
  • Operation 2002 depicts receiving a cue, the cue constructed via one or more operations associated with a media storage operation.
  • a cue the cue constructed via one or more operations associated with a media storage operation.
  • at least some data is received which is associated with a sample of video data taken from a particular client system.
  • the data may be associated with exactly the same patches of the client system as are defined by the ingest operation.
  • the data may be algorithmically processed to arrive at a hash value using the same operations as the ingest operation.
  • the same hashing operations as applied to the ingested frame will result in the same hash value as resulted from the hashing operations on the ingested frame.
  • the cue of operation 2002 represents data associated with a sample of video data from a particular client system.
  • a cue may be received via, for example, an HTTP request.
  • operation 2004 depicts referencing a number of most significant bits of the received cue to determine a database sector.
  • the same bits of the cue are examined as defined by the number of most significant bits used to reference a database sector during ingest. For example, if the first two bits of the hash value at ingest are used for storing the hash value at a particular database sector, the same first two bits of the cue associated with a sample of video data from a client system are used for addressing a particular database sector.
  • operation 2006 depicts returning at least one indication of at least one candidate from the database sector based at least partially on the received cue. For example, as shown in and/or described with respect to Figures 1 through 9, hash values which exactly match the cue, or are nearby the cue, are returned as one or more of suspects or candidates. Candidates may be returned within a particular percentage radius. Candidates may be returned according to a nearest neighbor algorithm or a modified nearest neighbor algorithm.
  • Figure 21 illustrates alternative embodiments of the example operational flow 2000 of Figure 20.
  • Figure 21 illustrates an example embodiment where operation 2002 may include at least one additional operation. Additional operations may include operation 2102, operation 2104, and/or operation 2106.
  • Operation 2102 illustrates receiving a cue associated with a sample of a video buffer of a client system, including at least receiving one or more indications related to an epoch time associated with the sample of the video buffer of the client system.
  • a cue may include, or be associated with, a time offset from an arbitrary time. The time offset may be computed from January 1, 1970, for example.
  • Operation 2104 illustrates receiving a cue, the cue associated with a sample of a video buffer of a client system, the cue at least partially determined by hashing at least some values associated with the video buffer.
  • patches associated with a video buffer may be reduced to a bit string via one or more mathematical operations or algorithms using one or more operands as constants, the constants pre-derived via operations described elsewhere herein with respect to hashing, for example.
  • Operation 2106 illustrates receiving a cue, the cue associated with a sample of a video buffer of a client system, the cue at least partially determined by hashing at least some values associated with the video buffer, the hashing based at least partially one or more of at least one operand or at least one algorithm also utilized in an associated media storage operation.
  • at least some data associated with a sample of a video buffer representing what is displayed by a television screen at a particular quantum of time is processed via operations utilized by the ingest process and/or in conjunction with data locations common to the ingest process and/or involving constant values for operands utilized by the ingest process.
  • the number of patches analyzed at ingest may also be utilized in providing a cue associated with a particular client system.
  • the size of pixel patches analyzed at ingest may also be utilized in providing a cue associated with a particular client system.
  • the same pre-derived static matrix used to more evenly distribute hash values at ingest may also be used during hashing of the data associated with a particular client system.
  • Figure 22 illustrates alternative embodiments of the example operational flow 2000 of Figure 20.
  • Figure 22 illustrates an example embodiment where operation 2002 may include at least one additional operation. Additional operations may include operation 2202, operation 2204, operation 2206, operation 2208, operation 2210, operation 2212, and/or operation 2214.
  • Operation 2202 illustrates receiving one or more indications of at least one content of a video buffer of a client system.
  • pixel values for red, green, and blue pixels at every pixel location at every pre-defined patch of the video buffer of the client system may be read, for every frame, or for every third frame, or for every tenth frame, or for every second, or at some other interval.
  • the indications (pixel values or other data) may be received by a widget on the television, by control logic on the television, by a system coupled with the media server, or elsewhere.
  • Operation 2204 illustrates determining, for at least one patch of the at least one content of the video buffer including at least one or more pixels of the at least one patch, an algorithmically-derived value of the one or more pixels of each patch. For example, as shown in and/or described with respect to Figures 1 through 9, pixel values for red, green, and blue pixels at every pixel location at every pre-defined patch of the video buffer of the client system may be averaged.
  • Operation 2206 illustrates subtracting a median point value from the mean value for each patch.
  • median point values at each patch established through analysis of ingested content are determined.
  • the median point values for each patch may, for example, be provided to the client system once determined by a system associated with the media database and ingest system.
  • the median point values may be updated from time to time (hourly, daily, monthly, yearly).
  • the median point values provided for hashing data associated with a video buffer of a client system may be the same median point values utilized to hash incoming content at ingest.
  • Operation 2208 illustrates transforming the values resulting from the subtraction.
  • values resulting from the subtraction are populated in a matrix and crossed with a pre-defined static matrix.
  • the dot-product operation crossing the two matrices may be conducted at the client system during a process of converting pixel patch data associated with a frame in a video buffer to a cue, such that a cue is sent in an HTTP request rather than the actual pixel patch data, resulting in a compact HTTP message.
  • the pre-defined static matrix may be provided to the client system in advance of the transform, and may be the same matrix as was produced to distribute hashed values at ingest more evenly.
  • the pre-defined static matrix may be updated at the client system from time to time.
  • patch data may be sent, with or without other metadata, from a client system (television, e.g.) to a different system for processing and/or hashing.
  • Operation 2210 illustrates constructing a hash value from the transformed values.
  • the values in the matrix resulting from crossing the matrix with values associated with the video buffer with the pre-derived static matrix may be reduced to bits, with a single bit replacing each 8-bit value in the matrix.
  • the constructed hash value may include a different number of bits for each value in the matrix.
  • the constructed hash value may have the same number of bits as the values in the matrix, or may be a direct representation of the values in the matrix.
  • Operation 2212 illustrates associating the cue at least partially with the constructed hash value.
  • the string of bits constructed from the transformed matrix may be a cue, or may associate the constructed string of bits with a time (such as an epoch time) to form a cue, or may associate other data such as an IP address or other identifier associated with the client television or a widget of the client television to form a cue.
  • the cue may include or otherwise be associated with any other metadata associated with audiovisual content at the client system.
  • Operation 2214 illustrates at least one of the determining 2204, subtracting 2206, transforming 2208, or constructing 2210 operations utilize one or more of at least one operand or at least one algorithm also utilized in an associated media storage operation.
  • one or more parameters including one or more of a definition of a number of pixel patches, a definition of a size of pixel patches, a pre-defined median value associated with pixel patches, or a pre-defined static matrix may be provided to a client TV, the one or more parameters also utilized by the ingest process such that operations applied to a sample from a video buffer will result in the same hash value that would result when that frame (e.g.
  • Figure 23 illustrates alternative embodiments of the example operational flow 2000 of Figure 20.
  • Figure 23 illustrates an example embodiment where operation 2006 may include at least one additional operation. Additional operations may include operation 2302, and/or operation 2304.
  • Operation 2302 illustrates returning at least one indication of at least one candidate from the database sector based at least partially on a probabilistic point location in equal balls (“PPLEB”) algorithm as a function of the received cue.
  • PPLEB probabilistic point location in equal balls
  • at least one of candidates or suspects representing path points close to a cue are returned from a media database constructed and/or modified via an ingest process.
  • Operation 2304 illustrates returning at least one indication of at least one candidate from the database sector based at least partially on the received cue, the at least one candidate being within a predetermined inverse percentage distribution radius of the received cue. For example, as shown in and/or described with respect to Figures 1 through 9, at least one of candidates or suspects associated with locality sensitive hashing related to at least one of a cue or a hash value are returned.
  • Figure 24 illustrates an operational flow 2400 representing example operations related to addressing a media database using distance associative hashing.
  • discussion and explanation may be provided with respect to the above-described examples of Figures 1 through 9, and/or with respect to other examples and contexts.
  • the operational flows may be executed in a number of other environments and contexts, and/or in modified versions of Figures 1 through 9.
  • the various operational flows are presented in the sequence(s) illustrated, it should be understood that the various operations may be performed in other orders than those which are illustrated, or may be performed concurrently.
  • Operation 2402 depicts receiving at least one indication of at least one candidate and at least one indication of at least one cue. For example, as shown in and/or described with respect to Figures 1 through 9, a hash value related to a video buffer of a client system, along with one or more associated candidates or suspects is determined.
  • operation 2404 depicts adding a token to a bin associated with at least one received candidate.
  • scoring of candidates is performed via tokens added to bins corresponding to candidates/suspects, the token being, for example, a value which is incremented each time a token is added.
  • operation 2406 depicts determining whether a number of tokens in a bin exceeds a value associated with a probability that a client system is displaying a particular video segment associated with at least one cue and, if the number of tokens in a bin exceeds a value associated with a probability that a client system is display a particular video segment associated with at least one cue, returning at least some data associated with the particular video segment based at least partially on the bin.
  • a determination of a particular video segment and particular offset of the video segment is probabilistically determined via the scoring associated with the bins.
  • Figure 25 illustrates alternative embodiments of the example operational flow 2400 of Figure 24.
  • Figure 25 illustrates an example embodiment where operation 2404 may include at least one additional operation 2502.
  • Operation 2502 illustrates adding a token to a time bin associated with at least one received candidate.
  • a data structure associated with a candidate/suspect may include an arbitrary time bin grouped by an arbitrary time.
  • Figure 26 illustrates alternative embodiments of the example operational flow 2400 of Figure 20.
  • Figure 26 illustrates an example embodiment where operation 2404 may include at least one additional operation. Additional operations may include operation 2602, and/or operation 2604. Further, operational flow 2400 may include at least one additional operation 2606.
  • Operation 2602 illustrates determining a relative time, including at least subtracting a candidate time associated with the at least one candidate from an arbitrary time associated with the at least one cue. For example, as shown in and/or described with respect to Figures 1 through 9, a time offset of a video segment associated with a candidate is subtracted from an arbitrary time associated with an epoch time related to the cue received from a client system (television, set-top box, or article, machine, or composition of matter displaying and/or providing and/or receiving video content).
  • a client system television, set-top box, or article, machine, or composition of matter displaying and/or providing and/or receiving video content.
  • Operation 2604 illustrates adding a token to a time bin associated with the candidate based at least partially on the determined relative time. For example, as shown in and/or described with respect to Figures 1 through 9, when a cue point associated with the client system matches or nearly matches a reference cue point associated with a media database, a token may be added to a bin, which may include incrementing a value associated with a bin or another means of tracking bin operations.
  • Operation 2606 illustrates removing one or more tokens from a time bin based at least partially on a time period elapsing.
  • a bin may be leaky such that data and/or tokens associated with old suspects/candidates may be release from the bin, which may include decrementing a value associated with a bin or another means of tracking bin operations.
  • pixel locations may relate to one or many colors and/or color spaces/models (e.g. red, blue, green; red, blue, green, and yellow; cyan, magenta, yellow, and black; a single pixel value uniquely identifying a color e.g. a 24-bit value associated with a pixel location; hue, saturation, brightness; etc.). Differing numbers of pixels in a patch may be used, and the patch does not have to be a square patch. Further, resolution of the video buffer of the client system may vary. Resolutions and/or color densities at the client system and the ingest system may vary.
  • the system may be operable with various raster resolutions, including but not limited to 1920 by 1080, 3840 by 2160, 1440 x 1080, 1366 x 768, or other resolutions. It is expected that over the next two decades, increases in pixel resolution of common programming, televisions, and/or client systems will occur; the same basic operations may be utilized although pixel patch number, size, sampling rate, or other aspects may vary. Further, an up-conversion, down-conversion, or other transformation operation associated with resolution and/or color density may occur and/or be interposed between other operations described herein. [00163]
  • Figure 27 illustrates an example system 2700 in which embodiments may be implemented.
  • the system 2700 includes one or more computing devices 2702.
  • the system 2700 also illustrates a fabric 2704 for facilitating communications among one or more computing devices and one or more client devices 2706.
  • the system 2700 also illustrates one or more client devices 2706.
  • the one or more client devices may be among the one or more computing devices.
  • the system 2700 also illustrates at least one non-transitory computer- readable medium 2708.
  • 2708 may include one or more instructions 2710 that, when executed on at least some of the one or more computing devices, cause at least some of the one or more computing devices to at least receive at least one stream of rasterized video; create at least one hash value associated with at least one sample of at least one received rasterized video stream; determine at least one database sector for storing a created at least one hash value; and store a created at least one hash value on at least one determined database sector.
  • the one or more instructions may be executed on a single computing device.
  • some portions of the one or more instructions may be executed by a first plurality of the one or more computing devices, while other portions of the one or more instructions may be executed by a second plurality of the one or more computing devices.
  • Figure 28 illustrates an example system 2800 in which embodiments may be implemented.
  • the system 2800 includes one or more computing devices 2802.
  • the system 2800 also illustrates a fabric 2804 for facilitating communications among one or more computing devices and one or more client devices 2806.
  • the system 2800 also illustrates one or more client devices 2806.
  • the one or more client devices may be among the one or more computing devices.
  • the system 2800 also illustrates at least one non-transitory computer- readable medium 2808.
  • 2808 may include one or more instructions 2810 that, when executed on at least some of the one or more computing devices, cause at least some of the one or more computing devices to at least receive one or more indications associated with at least one video buffer of at least one client system; determine a cue based at least partially on the at least one video buffer and at least one epoch time associated with the at least one video buffer, wherein one or more of at least one operand or at least one function associated with determining the cue is also utilized in an associated media storage operation; reference a number of most significant bits of a determined cue to determine a database sector; and return at least one indication of at least one candidate from a determined database sector based at least partially on a determined cue.
  • the one or more instructions may be executed on a single computing device. In other embodiments, some portions of the one or more instructions may be executed by a first plurality of the one or more computing devices, while other portions of the one or more instructions may be executed by a second plurality of the one or more computing devices.
  • Figure 29 illustrates an example system 2900 in which embodiments may be implemented.
  • the system 2900 includes one or more computing devices 2902.
  • the system 2900 also illustrates a fabric 2904 for facilitating communications among one or more computing devices and one or more client devices 2906.
  • the system 2900 also illustrates one or more client devices 2906.
  • the one or more client devices may be among the one or more computing devices.
  • the system 2900 also illustrates at least one non-transitory computer- readable medium 2908.
  • 2908 may include one or more instructions 2910 that, when executed on at least some of the one or more computing devices, cause at least some of the one or more computing devices to at least receive at least one indication of at least one candidate and at least one indication of at least one cue; add a token to a bin associated with at least one received candidate; and determine whether a number of tokens in a bin exceeds a value associated with a probability that a client system is receiving a particular video segment associated with at least one received cue and, if the number of tokens in a bin exceeds a value associated with a probability that a client system is receiving a particular video segment associated with at least one received cue, returning at least some data associated with the particular video segment based at least partially on the bin.
  • the one or more instructions may be executed on a single computing device. In other embodiments, some portions of the one or more instructions may be executed by a first plurality of the one or more computing devices, while other portions of the one or more instructions may be executed by a second plurality of the one or more computing devices.
  • Certain aspects of the present invention include process steps and instructions described herein in the form of an algorithm. It should be noted that the process steps and instructions of the present invention could be embodied in software, firmware or hardware, and when embodied in software, could be downloaded to reside on and be operated from different platforms used by real-time network operating systems.
  • the present invention also relates to an apparatus for performing the operations herein.
  • This apparatus may be specially constructed for the required purposes, or it may comprise a general-purpose computer selectively activated or reconfigured by a computer program stored in the computer.
  • a computer program may be stored in a computer readable storage medium, such as, but is not limited to, any type of disk including floppy disks, optical disks, CD-ROMs, magnetic-optical disks, read-only memories (ROMs), random access memories (RAMs), EPROMs, EEPROMs, magnetic or optical cards, application specific integrated circuits (ASICs), or any type of media suitable for storing electronic instructions, and each coupled to a computer system bus.
  • computers or computing means referred to in the specification may include a single processor or may employ multiple-processor designs for increased computing capability.
  • Embodiments of the subject matter described in this specification can be implemented as one or more computer program products, i.e., one or more modules of computer program instructions encoded on a tangible program carrier for execution by, or to control the operation of, data processing apparatus.
  • the computer readable medium can be a machine readable storage device, a machine readable storage substrate, a memory device, or a combination of one or more of them.
  • a computer program (also known as a program, software, software application, script, or code) can be written in any form of programming language, including compiled or interpreted languages, or declarative or procedural languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.
  • a computer program does not necessarily correspond to a file in a file system.
  • a program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code).
  • a computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a suitable communication network.
  • the essential elements of a computer are a processor for performing instructions and one or more memory devices for storing instructions and data.
  • a computer will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magneto optical disks, or optical disks.
  • mass storage devices for storing data, e.g., magnetic, magneto optical disks, or optical disks.
  • processors suitable for the execution of a computer program include, by way of example only and without limitation, both general and special purpose microprocessors, and any one or more processors of any kind of digital computer.
  • a processor will receive instructions and data from a read only memory or a random access memory or both.
  • a computer having a display device, e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor, for displaying information to the user and a keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computer.
  • a display device e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor
  • a keyboard and a pointing device e.g., a mouse or a trackball
  • Other kinds of devices can be used to provide for interaction with a user as well.
  • feedback provided to the user can be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback
  • input from the user can be received in any form, including acoustic, speech, or tactile input.
  • Embodiments of the subject matter described in this specification can be implemented in a computing system that includes back end component(s) including one or more data servers, or that includes one or more middleware components such as application servers, or that includes a front end component such as a client computer having a graphical user interface or a Web browser through which a user or administrator can interact with some implementations of the subject matter described is this specification, or any combination of one or more such back end, middleware, or front end components.
  • the components of the system can be interconnected by any form or medium of digital data communication, such as a communication network.
  • the computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client server relationship to each other.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Library & Information Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Human Computer Interaction (AREA)
  • Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Computer Security & Cryptography (AREA)
  • Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Television Signal Processing For Recording (AREA)

Abstract

A system, method and computer program utilize a distance associative hashing algorithmic means to provide a highly efficient means to rapidly address a large database. The indexing means can be readily subdivided into a plurality of independently-addressable segments where each such segment can address a portion of related data of the database where the subdivided indexes of said portions reside entirely in the main memory of each of a multiplicity of server means. The resulting cluster of server means, each hosting an addressable sector of a larger database of searchable audio or video information, provides a significant improvement in the latency and scalability of an Automatic Content Recognition system, among other uses.

Description

SYSTEMS AND METHODS FOR ADDRESSING A MEDIA DATABASE USING DISTANCE ASSOCIATIVE HASHING
PRIORITY CLAIM
[0001] This application constitutes a continuation-in-part of U.S. Patent Application No. 12/788,721, entitled "METHODS FOR IDENTIFYING VIDEO SEGMENTS AND DISPLAYING CONTEXTUAL TARGETED CONTENT ON A CONNECTED TELEVISION," filed May 27, 2010, and issued November 6, 2013 as U.S. Patent No. 8,595,781, that application being a non-provisional application claiming priority from U.S. Provisional Patent Application No. 61/182,334, entitled "SYSTEM FOR PROCESSING CONTENT INFORMATION IN A TELEVIDEO SIGNAL," filed May 29, 2009 and being a non- provisional application claiming priority from U.S. Provisional Patent Application No. 61/290,714, entitled "CONTEXTUAL TARGETING BASED ON DATA RECEIVED FROM A TELEVISION SYSTEM," filed December 29, 2009; this application further constitutes a continuation-in-part of U.S. Patent Application No. 12/788,748, entitled "METHODS FOR DISPLAYING CONTEXTUALLY TARGETED CONTENT ON A CONNECTED TELEVISION," filed May 27, 2010; this application further constitutes a continuation-in-part of
U.S. Patent Application No. 14/089,003, entitled " ," filed November 25,
2013; this application further constitutes a continuation-in-part of U.S. Patent Application No. **/***,***, entitled "SYSTEMS AND METHODS FOR IDENTIFYING VIDEO SEGMENTS FOR DISPLAYING CONTEXTUALLY RELEVANT CONTENT," filed March 17, 2014; this application further constitutes a continuation-in-part of U.S. Patent Application No. **/*** ***^ entitled "SYSTEMS AND METHODS FOR REAL-TIME TELEVISION AD DETECTION USING AN AUTOMATED CONTENT RECOGNITION DATABASE," filed March 17, 2014; this application further constitutes a continuation-in-part of U.S. Patent Application No. **/***,***, entitled "SYSTEMS AND METHODS FOR ON-SCREEN GRAPHICS DETECTION," filed March 17, 2014; this application further constitutes a continuation-in-part of U.S. Patent Application No. **/*** ***, entitled "SYSTEMS AND METHODS FOR IMPROVING SERVER AND CLIENT PERFORMANCE IN FINGERPRINT ACR SYSTEMS," filed March 17, 2014; this application further constitutes a continuation-in-part of U.S. Patent Application No. **/*** ***, entitled "SYSTEMS AND METHODS FOR MULTI- BROADCAST DIFFERENTIATION," filed March 17, 2014; and this application further constitutes a non-provisional application of U.S. Provisional Patent Application No. 61/791,578, entitled "SYSTEMS AND METHODS FOR IDENTIFYING VIDEO SEGMENTS BEING DISPLAYED ON REMOTELY LOCATED TELEVISIONS," filed March 15, 2013. The foregoing applications are either currently co-pending or are applications of which a currently co-pending application is entitled to the benefit of the filing date.
FIELD OF THE INVENTION
[0002] This invention generally relates to the matching of unknown media data, such as video or audio segments, against a massive database of reference media files.
BACKGROUND
[0003] Systems for automatic content recognition (ACR) of audio or video media are well known to persons skilled in the art. However, such ACR systems pose many technical challenges, including managing potentially very large databases of encoded audio or video information as well as managing large indices needed for addressing information in said databases.
[0004] Also well known to those skilled in the art, is that large database indices such as may be used by this invention, can be generated using certain hashing functions. Another method of addressing a database might be by applying binary tree structures also known as b- trees. Both methods are commonly utilized in data management systems.
[0005] Whatever the method employed to index a large database, said index is often too large to reside in its entirety in the main memory of a computer server as used in a typical ACR system. When said database cannot fit entirely in the memory of a computer system, it is typically stored on magnetic disk storage and parts of said database are then read into memory in blocks corresponding to the index value providing the address. Said means of recalling partial database information is also known to one skilled in the art as "paging" which is a process common to many different computer software systems.
[0006] The present invention is an extension of the invention referenced above and is a system and method for matching unknown digital media such as television programing to a database of known media using a signal processing means employing a modified path pursuit algorithm, as described in the first invention.
[0007] Another novel aspect of the system and method as disclosed herein is its distance associative hash indexing means which can be subdivided into a plurality of independently addressable segments where each of said segments can address a portion of the database each of which can reside in its entirety in the main memory of a server means. The resulting cluster of servers of the indexing means each hosts a sector of the index addressing associated data of a larger database of searchable audio or video information. This indexing means of the invention results in a significant improvement in the speed and accuracy of the ACR system so enabled as to identify unknown media even when the television display is showing content where a user is changing channels, rewinding, fast-forwarding or even pausing video from a digital video recorder.
SUMMARY
[0008] In some embodiments, an exemplary method related to addressing a media database using distance associative hashing may include receiving one or more indications of a sample of a video segment; determining, for at least one patch of the sample of a video segment including at least one or more pixels of the at least one patch, an algorithmically-derived value of the one or more pixels of each patch; subtracting a median point value established for each patch from the mean value for each patch; transforming the values resulting from the subtraction using a function pre-derived to distribute the values evenly; constructing a hash value from the transformed values; referencing a number of most significant bits of the constructed hash value to determine a database sector; and storing at least the hash value on the determined database sector.
[0009] In some embodiments, at least one of the receiving, determining, subtracting, transforming, constructing, referencing, or storing of the foregoing exemplary method is at least partially implemented using one or more processing devices. In some embodiments of the foregoing exemplary method, receiving one or more indications of a sample of a video segment may include receiving one or more indications of at least one of a frame or a still image. In some embodiments of the foregoing exemplary method, receiving one or more indications of a sample of a video segment may include receiving one or more indications of a sample of a video segment, the one or more indications of a sample of a video segment associated with at least one indication of a channel, at least one indication of a video segment, and at least one indication of a timecode offset from the beginning of the video segment.
[0010] In some embodiments of the foregoing exemplary method, determining, for at least one patch of the sample of a video segment including at least one or more pixels of the at least one patch, an algorithmically-derived value of the one or more pixels of each patch includes at least determining, for at least one patch of the sample of a video segment including at least one or more pixels of the at least one patch, a mean value of the one or more pixels of each patch. In some embodiments of the foregoing exemplary method, subtracting a median point value established for each patch from the mean value for each patch may include subtracting a median point value established for each patch from the mean value for each patch, the median point value established for each patch having been previously determined utilizing data from each patch for a plurality of channels over at least one period of time.
[0011] In some embodiments of the foregoing exemplary method, transforming the values resulting from the subtraction using a function pre-derived to distribute the values evenly may include forming a variable matrix including at least the values resulting from the subtraction; obtaining a static matrix which, when crossed with the variable matrix, will more evenly distribute the transformed values; and computing a dot product of the variable matrix and the static matrix, the dot product including at least the more evenly-distributed transformed values. In some embodiments of the foregoing exemplary method, obtaining a static matrix which, when crossed with the variable matrix, will more evenly distribute the transformed values may include determining, using locality-sensitive hashing at least partially based on one or more previously obtained hash values, a static matrix which, when crossed with a variable matrix, will more evenly distribute the transformed values of the variable matrix.
[0012] In some embodiments of the foregoing exemplary method, constructing a hash value from the transformed values may include constructing a hash value from the transformed values, including at least reducing the fidelity of the transformed values via reducing each transformed value to a binary representation. In some embodiments of the foregoing exemplary method, reducing the fidelity of the transformed values via reducing each transformed value to a binary representation may include determining for each transformed value whether the transformed value is a positive number and, if the transformed value is a positive number, assigning a one to the hash value and otherwise assigning a zero to the hash value.
[0013] In some embodiments of the foregoing exemplary method, referencing a number of most significant bits of the constructed hash value to determine a database sector may include referencing a number of most significant bits of the constructed hash value to determine a database server, wherein the number of most significant bits is pre-determined to address a plurality of database servers, wherein a number of database servers associated with the number of most significant bits is established to enable at least one index associated with a database sector to reside entirely in memory of a corresponding database server. In some embodiments of the foregoing exemplary method, storing at least the hash value on the determined database sector may include storing at least the hash value on the determined database sector, including at least storing at least one indication of a channel, at least one indication of a video segment, and at least one indication of a timecode offset from the beginning of the video segment at a database location at least partially based on the hash value.
[0014] In one or more alternative embodiments of the foregoing exemplary method, related systems include but are not limited to circuitry and/or programming for effecting the herein-referenced method embodiments; the circuitry and/or programming can be virtually any combination of hardware, software, and/or firmware configured to effect the herein-referenced method aspects depending upon the design choices of the system designer.
[0015] In a different embodiment, an exemplary method related to addressing a media database using distance associative hashing may include receiving a cue, the cue constructed via one or more operations associated with a media storage operation; referencing a number of most significant bits of the received cue to determine a database sector; and returning at least one indication of at least one candidate from the database sector based at least partially on the received cue.
[0016] In some embodiments of the foregoing exemplary method, receiving a cue, the cue constructed via one or more operations associated with a media storage operation may include receiving a cue associated with a sample of a video buffer of a client system, including at least receiving one or more indications related to an epoch time associated with the sample of the video buffer of the client system. In some embodiments of the foregoing exemplary method, receiving a cue, the cue constructed via one or more operations associated with a media storage operation may include receiving a cue, the cue associated with a sample of a video buffer of a client system, the cue at least partially determined by hashing at least some values associated with the video buffer.
[0017] In some embodiments of the foregoing exemplary method, receiving a cue, the cue associated with a sample of a video buffer of a client system, the cue at least partially determined by hashing at least some values associated with the video buffer may include receiving a cue, the cue associated with a sample of a video buffer of a client system, the cue at least partially determined by hashing at least some values associated with the video buffer, the hashing based at least partially one or more of at least one operand or at least one algorithm also utilized in an associated media storage operation. In some embodiments of the foregoing exemplary method, receiving a cue, the cue constructed via one or more operations associated with a media storage operation may include receiving a cue, the cue determined via one or more operations including at least receiving one or more indications of at least one content of a video buffer of a client system; determining, for at least one patch of the at least one content of the video buffer including at least one or more pixels of the at least one patch, an algorithmically- derived value of the one or more pixels of each patch; subtracting a median point value from the mean value for each patch; transforming the values resulting from the subtraction; constructing a hash value from the transformed values; and associating the cue at least partially with the constructed hash value, wherein at least one of the determining, subtracting, transforming, or constructing operations utilize one or more of at least one operand or at least one algorithm also utilized in an associated media storage operation.
[0018] In some embodiments of the foregoing exemplary method, returning at least one indication of at least one candidate from the database sector based at least partially on the received cue may include returning at least one indication of at least one candidate from the database sector based at least partially on a probabilistic point location in equal balls ("PPLEB") algorithm as a function of the received cue. In some embodiments of the foregoing exemplary method, returning at least one indication of at least one candidate from the database sector based at least partially on the received cue may include returning at least one indication of at least one candidate from the database sector based at least partially on the received cue, the at least one candidate being within a predetermined inverse percentage distribution radius of the received cue.
[0019] In one or more alternative embodiments of the foregoing exemplary method, related systems include but are not limited to circuitry and/or programming for effecting the herein-referenced method embodiments; the circuitry and/or programming can be virtually any combination of hardware, software, and/or firmware configured to effect the herein-referenced method aspects depending upon the design choices of the system designer.
[0020] In a different embodiment, an exemplary method related to addressing a media database using distance associative hashing may include receiving at least one indication of at least one candidate and at least one indication of at least one cue; adding a token to a bin associated with at least one received candidate; and determining whether a number of tokens in a bin exceeds a value associated with a probability that a client system is displaying a particular video segment associated with at least one cue and, if the number of tokens in a bin exceeds a value associated with a probability that a client system is display a particular video segment associated with at least one cue, returning at least some data associated with the particular video segment based at least partially on the bin. [0021] In some embodiments of the foregoing exemplary method, adding a token to a bin associated with at least one received candidate may include adding a token to a time bin associated with at least one received candidate. In some embodiments of the foregoing exemplary method, adding a token to a bin associated with at least one received candidate may include determining a relative time, including at least subtracting a candidate time associated with the at least one candidate from an arbitrary time associated with the at least one cue; and adding a token to a time bin associated with the candidate based at least partially on the determined relative time. In some embodiments of the foregoing exemplary method, the method may include removing one or more tokens from a time bin based at least partially on a time period elapsing.
[0022] In one or more alternative embodiments of the foregoing exemplary method, related systems include but are not limited to circuitry and/or programming for effecting the herein-referenced method embodiments; the circuitry and/or programming can be virtually any combination of hardware, software, and/or firmware configured to effect the herein-referenced method aspects depending upon the design choices of the system designer.
[0023] In a different embodiment, an exemplary system related to addressing a media database using distance associative hashing may include, but is not limited to, one or more computing devices; and one or more instructions that, when executed on at least some of the one or more computing devices, cause at least some of the one or more computing devices to at least receive at least one stream of rasterized video; create at least one hash value associated with at least one sample of at least one received rasterized video stream; determine at least one database sector for storing a created at least one hash value; and store a created at least one hash value on at least one determined database sector.
[0024] In a different embodiment, an exemplary system related to addressing a media database using distance associative hashing may include, but is not limited to, one or more computing devices; and one or more instructions that, when executed on at least some of the one or more computing devices, cause at least some of the one or more computing devices to at least receive one or more indications associated with at least one video buffer of at least one client system; determine a cue based at least partially on the at least one video buffer and at least one epoch time associated with the at least one video buffer, wherein one or more of at least one operand or at least one function associated with determining the cue is also utilized in an associated media storage operation; reference a number of most significant bits of a determined cue to determine a database sector; and return at least one indication of at least one candidate from a determined database sector based at least partially on a determined cue.
[0025] In a different embodiment, an exemplary system related to addressing a media database using distance associative hashing may include, but is not limited to, one or more computing devices; and one or more instructions that, when executed on at least some of the one or more computing devices, cause at least some of the one or more computing devices to at least receive at least one indication of at least one candidate and at least one indication of at least one cue; add a token to a bin associated with at least one received candidate; and determine whether a number of tokens in a bin exceeds a value associated with a probability that a client system is receiving a particular video segment associated with at least one received cue and, if the number of tokens in a bin exceeds a value associated with a probability that a client system is receiving a particular video segment associated with at least one received cue, returning at least some data associated with the particular video segment based at least partially on the bin.
[0026] In addition to the foregoing, various other methods, systems and/or program product embodiments are set forth and described in the teachings such as the text (e.g., claims, drawings and/or the detailed description) and/or drawings of the present disclosure.
[0027] The foregoing is a summary and thus contains, by necessity, simplifications, generalizations and omissions of detail; consequently, those skilled in the art will appreciate that the summary is illustrative only and is NOT intended to be in any way limiting. Other aspects, embodiments, features and advantages of the device and/or processes and/or other subject matter described herein will become apparent in the teachings set forth herein.
BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Certain embodiments of the present invention are described in detail below with reference to the following drawings:
[0029] Figure 1 illustrates the construction of a sectored video matching database as taught by this invention which begins with initial video ingest or capture process which is then continuously updated. A television display system 101 and its corresponding television display memory buffer 103 are shown for a potential embodiment of the system. The allocation of pixel patches 102 and a calculation of a value 105, using certain algorithmic means known to those skilled in the art, is made for each pixel patch and a resulting data structure is created and then time-stamped make a "cue" 106 which may also have additional metadata associated with it.
[0030] Figure 2 illustrates the processing of the cue data 201 and the generation of the hash index 202 using the distance associative hashing process, further illustrating the sectored addressing scheme 203 to store data in related groups (buckets) 206.
[0031] Figure 3: illustrates the real-time capture of unknown television content for recognition from a connected television monitor or the like 301. A pixel patch is defined as typically a square pixel area of the video buffer 303 with dimensions of perhaps ten pixels by ten rows of pixels 304, however, any reasonable shape and dimension may be used. The number of pixel patch positions can be any number between ten and fifty locations within said video buffer and is processed 305 to send cue data 306 to the central server means.
[0032] Figure 4: illustrates the extraction of candidate cue values 401 from the reference (matching) database bucket 404 and supplying said cue values 403 to the path pursuit content matching process 402 as taught in the first invention referenced above.
[0033] Figure 5: illustrates the data structure of bins which hold tokens for scoring candidate values from the matching database. Said bins are "leaky" and tokens expire over time as the search process progresses through time.
[0034] Figure 6: illustrates a typical memory paging scheme as taught by prior art for accessing large databases.
[0035] Figure 7 illustrates the creation of a hash value involving several steps beginning with computing the median value of each of the multiplicity of points which make up the samples from a frame of video.
[0036] Figure 8 illustrates how the hash value is computed.
[0037] Figure 9 illustrates the beneficial results of using the median values of a pixel location as part of the process of computing the hash values.
[0038] Figure 9a illustrates the problem of not using a media value when partitioning a multi-dimensional dataset.
[0039] Figure 9b illustrates the benefit of finding a media value of a dataset. [0040] Figure 10 illustrates an operational flow representing example operations related to addressing a media database using distance associative hashing.
[0041] Figure 11 illustrates an alternative embodiment of the operational flow of Figure 10.
[0042] Figure 12 illustrates an alternative embodiment of the operational flow of Figure 10.
[0043] Figure 13 illustrates an alternative embodiment of the operational flow of Figure 10.
[0044] Figure 14 illustrates an alternative embodiment of the operational flow of Figure 10.
[0045] Figure 15 illustrates an alternative embodiment of the operational flow of Figure 10.
[0046] Figure 16 illustrates an alternative embodiment of the operational flow of Figure 10.
[0047] Figure 17 illustrates an alternative embodiment of the operational flow of Figure 10.
[0048] Figure 18 illustrates an alternative embodiment of the operational flow of Figure 10.
[0049] Figure 19 illustrates an alternative embodiment of the operational flow of Figure 10.
[0050] Figure 20 illustrates a different operational flow representing example operations related to addressing a media database using distance associative hashing.
[0051] Figure 21 illustrates an alternative embodiment of the operational flow of Figure 20.
[0052] Figure 22 illustrates an alternative embodiment of the operational flow of Figure 20.
[0053] Figure 23 illustrates an alternative embodiment of the operational flow of Figure 20.
[0054] Figure 24 illustrates another operational flow representing example operations related to addressing a media database using distance associative hashing. [0055] Figure 25 illustrates an alternative embodiment of the operational flow of Figure 24.
[0056] Figure 26 illustrates an alternative embodiment of the operational flow of Figure 24.
[0057] Figure 27 illustrates a system related to addressing a media database using distance associative hashing.
[0058] Figure 28 illustrates another system related to addressing a media database using distance associative hashing.
[0059] Figure 29 illustrates yet another system related to addressing a media database using distance associative hashing.
DETAILED DESCRIPTION
[0060] The first invention which relates to this invention is a system and method of matching unknown video to a database of known video using a novel signal processing means employing a modified path pursuit algorithm, among other means, as described in the aforementioned publication.
[0061] A novel means of the new invention is its Distance Associated Hashing with its attendant provision of utilizing a sectored-index database access. Said indexing means provides a highly computationally-efficient means for matching an unknown media segment to a reference database of known media, such as audio or video content.
[0062] This indexing means of the invention results in a significant improvement in the speed and accuracy of the ACR system so enabled as to track the identity of media even when the television display is showing content where a user is changing channels, rewinding, fast- forwarding or even pausing video from a digital video recorder.
[0063] Both the building, updating, and the subsequent accessing of the media matching database will describe a system capable of generating and addressing a sectored database such that the database sectors can each reside in the main memory of a respective multiplicity of server means without resorting to a paging means within each of the respective server means. This collective means of addressing a sectored database through locality sensitive hashing provides a significant improvement in efficiency of operation. [0064] The construction of a sectored video matching database begins with the process as illustrated in Figure 1. A television system 101 decodes a television signal and places the contents of each frame of video into a video frame buffer in preparation for the display or further processing of pixel information of the frame of video. Said television system can be any television decoding system that can decode a television signal whether from a baseband or modulated television source and fill a video frame buffer with the decoded RGB values at the respective frame size as specified by the video signal. Such systems are well known to one skilled in the art.
[0065] The system of the invention first builds and then continuously updates a reference database of television programming fingerprints described in the original application as cues or cue values. For purposes of building said reference database of video cues, the invention performs the acquisition of one or more patches of video 102 which are read from the video frame buffer 103. Said video patches can be any arbitrary shape or pattern but for the purposes of this example shall be 10 pixels horizontally by 10 pixels vertically. Also for the sake of this example, assume that there are 25 pixel patch positions within the video frame buffer that are evenly distributed within the boundaries of said buffer, though they do not have to be evenly distributed. Each pixel shall consist of a red, a green and a blue value, 104, typically represented by an eight bit binary value for each color for a total of 24 bits or three bytes per patch location.
[0066] This composite data structure is populated with the average pixel values from a number of pixel patch positions from the video buffer. A pixel patch is defined as a typically square pixel area of the video buffer with dimensions of perhaps ten pixels by ten rows of pixels 304. The number of pixel patch positions might typically be between ten and fifty locations within the video buffer.
[0067] The average pixel values 305 are assembled with a time code 306 referencing the "epoch time" from the processor means of the television system. Epoch time is defined as the time in fractions of a second that have elapsed since midnight, January 1,1970 which is an accepted convention in computing systems, particularly with Unix (or Linux)-based systems.
[0068] In addition, metadata may be included and together a data structure 106 is defined called a tagged fingerprint, "cue", or a "point", as taught in the original patent application. Such metadata attributes might be derived from closed-captioning data from the currently displaying video program or it could be keywords extracted by means of a speech recognition system operating within the processor means of the television system which converts audio from the respective television program into text information. Said textual information may then be searched for relevant key words or sent in its entirety as part of the cue data structure to the central server means for further processing.
[0069] The cue records 201 are passed in Figure 2 to a hash function 202 that generates a hash value 203 using a locality sensitive hashing algorithm based on Probabilistic Point Location in Equal Balls algorithm (PPLEB). This hash value is computed from the averaged pixel values from the cue record (fingerprint) 207 and the process associates 206 data with like values into groups called buckets.
[0070] The ten by ten pixel patch 302 shown in this particular example would have one hundred pixels and is mathematically averaged resulting in a mean pixel value 305 for red, green and blue values, respectively. Alternatively, any averaging function can be used in place of a simple mean.
[0071] A plurality of such pixel patches are extracted from the video frame. If, by way of example, 25 such pixel patches are extracted from the video frame, then the result will be a point representing a position in a 75-dimension space. The skilled person will know that such a large search space could require extensive computing resources to later locate, even approximately, said value in combination with the other 74 values representing one frame of video.
[0072] It is an advantage of system and method of distance associated hashing described herein to reduce the computational load and improve accuracy of matching unknown video frames to a known video frame database.
[0073] The creation of the hash value involves several steps beginning with computing an algorithmically-derived value of each point as shown in Figure 7, 701 to 775. One useful means of algorithmically deriving said median value is found by summing each point of every frame of every program stream or channel of video maintained by the matching database over perhaps a 24 hour period. The median of each point is found from the summation process. The next step in deriving the final hash value is to subtract the mean value from the point value of each respective point, row 801 minus row 802 equals row 803. The result is a plus or minus values to which a pre-derived hashing function is applied. Typically, the result of the point values minus the mean values of the respective points, are arranged in a matrix to which a dot- product is calculated using a similar matrix constituting the pre-derived hash value (or key). The result of the dot product of the two matrices is then further transformed to a one or zero value based on the sign of the product matrix element. Typically the skilled person would set positive values to one and negative values to zero.
[0074] The resulting hash value points to more or less evenly distributed values across the data storage area. The hash value 203 can be further divided, Figure 2, such that the 'n' most significant bits 205 addresses one of the 2n (2An) sectors of the database. The remaining bits 206 address individual 'buckets' of the addressed sector of the database, which will be described in more detail later.
[0075] The division point of the hash value that defines the individual sector address space is calculated such that the data of the database sector's index fits within the memory confines of the processor systems of said memory sector. Otherwise, said database would be subject to paging which would diminish the effectiveness of this process.
[0076] To contrast the system and method taught by the present invention with that which is known to those skilled in the relevant prior art, Figure 6 illustrates a typical paging approach. In Figure 6, assume the example system is attempting to match unknown data to a database of known data. An index 602 is used to address only the portion of the data 605 that can fit in the main memory of the CPU 606. This data is searched and, if results are negative, then another segment of data is fetched into main memory 603 and searching continues.
[0077] Such accessing means using paging are common but considerably reduce the efficiency of a computer system. In fact, such an approach could not be uses with ACR systems searching large media database as the read / write speed of magnetic hard drives is insufficient to keep up with the task. Many different algorithmic approaches have been developed over the years to address this issue of splitting the search into smaller parts and allocating smaller searches to multiple computer server systems.
[0078] A notable example might be the considerable Google search engine. The skilled person knows this system to be one of the largest computer systems built to date. The speed and accuracy of the Google search process is remarkable. However, the Google search means is considerably different and not at all applicable to matching unknown media to a database of known media even if the two databases were the same very large size. This is because the Google search means employs the map-reduce algorithm which is designed for searching large databases of essentially unassociated data. While an advance over paging system, map-reduce is a computationally-intense process which also requires significant data communications bandwidth between the participating computer systems. In contrast, this invention is efficient in the use of processing and communications resources.
[0079] In this invention, the distance associative hashing function provides a means to address a database in sectors such that the data of said addressing means fits in the main memory of an individual server means of group of servers. Said grouping is accomplished by grouping the data related by distance in a multi-dimensional array into the same sector using the distance associative hashing step as a means to achieve said grouping. The sector identification for addressing a data element is calculated from the hash index generated from said process by extracting a subset of the total bits of said hash function and using said subset to address the desired sector in which to store data in the reference database.
[0080] In this manner, the hash index subset is the address of the sector that contains the distance associated hash values, called buckets in the first invention. The remainder of the hash address is then used to address a bucket of the sector for storing the new data. Alternatively, the sector address can be found by means of re-hashing the first hash value.
[0081] This system and method of database addressing by means of multiple hash- indexing steps produces a highly efficient database accessing scheme with significant performance benefits and increased efficiency over traditional methods of database access as described above.
[0082] The distance associative hashing provides a means to address a very complex (multi-dimensional) database quickly by finding data that is not an exact match but rather is within a predetermined radius (distance associative) of the value sought. Importantly, sometimes this addressing means will result in no match at all. Where a business-oriented database cannot tolerate inaccuracy, a media matching system can readily tolerate missed matches and will simply continue the matching process upon the arrival of the next data received and taught in the first patent. Data arrival from the unknown source that is to be determined by the ACR system is periodic, of course, but can be commanded by the system of the invention to arrive at differing intervals based on the requirement for accuracy or by requirements imposed by the state of the system such as when the system might be nearing overload and the sending clients are then commanded to send a lower sample rate. A typical data reception rate might be 1/10 second intervals, for example.
[0083] For the reference media database, the group of pixel values are derived from every frame of video from every video source that is to be part of the reference database. The group of pixel values and are then appended with the broadcast time of the video program as well as with certain metadata, which is information about the program such as the content identification (ID), title of the program, actors name, time of airing, short synopsis, etc. Said metadata is generally acquired from commercial electronic program guide sources.
[0084] Said array of processed pixel values with the addition of the timecode plus the metadata are then stored in the reference database and the address of said stored data is then added to the hash index at the respective hash value and sector ID value. In addition, a second database index is built and maintained by using the content ID from the metadata as another means for addressing the reference database.
[0085] The process of building and continuously updating the database is continuous and the number of days of data maintained by the database is based on the needs of the user but for example might range from one day to one month.
[0086] The process of identifying an unknown video segment from data received from a multiplicity of client devices begins with a procedure similar to that used above for building the reference database. In Figure 3, this procedure involves a television monitor 301, such as a popular flat-screen HDTV typically of the type known as the smart TV wherein the TV contains a processing means with memory capable of executing application programs similar to the type found on common smartphones of today. The system of the invention samples regions 302 of a video frame buffer 301 in typically a multiplicity of places. Said samples are of an identical size, shape and position to the pixel patches used in the process of building the reference database. Each of the collected pixel patches is then algorithmically processed to produce a computed value for the red, green and blue values of each patch in a manner identical to the method used to create the reference database. [0087] Said system of the invention then calculates a distance associative hash index of the collected mean values identical to the content ingest function described above. The resulting sector identification (ID) value is extracting as a subset of the total bits of the hash index also identically to the ingest system described above. The remainder of the hash index is used to address the desired sector in which to search for all candidate (potential) matches belonging to the same bucket as the unknown data point.
[0088] Optionally, if a good guess of a match (a successful match) was available from the process above, the system of the invention will also collect candidates from the database responsible for said sector belonging to the potential content ID, using the content ID index, created during the ingest process as described above, to address reference cues around time radius r' of the timestamp (of the successfully matched candidate). Duplicate candidates are next removed as well as candidates that are too far from the unknown point by radius r, as taught in the first patent.
[0089] In order to test for a match of an unknown video segment against a reference database of known video data, assume the list of candidates from the previous step where each candidate (i.e. each possible match) has associated with it the following data items: content ID, media time, inverse percentage distribution radius which is calculated as the distance from the current unknown point (from unknown video stream) where 100% represents the exact value of the unknown point and 0% is a value beyond the radius r (distribution) from the unknown point.
[0090] Each matching candidate 501 is assigned a data structure 502 in the memory of the matching system of the invention. The data structure consists of, among other things, arbitrary time bins grouped by some arbitrary amount (e.g. approximately one second). For the sake of example, assume said data structure consists of one hundred bins representing ten seconds of video cue points. The bins are generally not equally spaced in time.
[0091] For each candidate found in the reference (match) database: first, a relative time is calculated by subtracting candidate time from the arbitrary time of the unknown video. Candidate time is the time of broadcast of each video cue associated with the candidate during the reference program airing.
[0092] The arbitrary time of the unknown video came from the internally generated epoch time of the television monitor from the application of the invention operating in the memory of said television or in a set-top box attached to said television and sent by said application to the central server means of the invention with the sampled video cue points. Epoch time is well known to the skilled person and is typically employed in computer systems. Said time is calculated as the current number of units of time since January 1 of 1970.
[0093] If, for example, the time difference between the arbitrary time from the television (in the home) and the true media time is 100 seconds, then the relative time of the actually matching candidates should be close to that value. Likewise, candidates that are not a good match are not likely to have relative times close to the 100 seconds of this example.
[0094] In the candidate data structure, when a cue point of the unknown video matches a reference cue point, the system of the invention adds a token to the respective bin of the candidate data structure. Said system then repeats the process for the next candidate as described in the previous paragraph.
[0095] Another, and important, step for the scoring of results is to apply time discounting to all bins. This is a relatively simple process that decrements the value in all bins by a small amount for each cycle of time. The skilled person would recognize this as a "leaky bucket" method of scoring. By definition, bins that are no longer being filled by means of matching cue points will ultimately decrement to zero over a number of cycles of said process. Also, bins that are filled slowly by random noise in the system will likewise be decremented. Hence, time discounting ultimate clears bins that are filled by false-positive matches and random noise. The skilled person would also clearly see that without said time discount binning, all bins would eventually fill to capacity and no results could be obtained from the process.
[0096] Said time discounting also decrements to zero any bins with levels, such as 503, that are above the matching threshold 504 when the video stream from the client television monitor is in any way changed by any of the following: changing channels, rewind, fast forward, pausing video, etc.
[0097] If any bin of the candidate data structure is above a certain threshold 504, such as bin 503, then the content is declared a match. Further means to qualify a match might include testing for contiguous matches of the candidate segment for greater that a determined number of seconds (e.g. three seconds). [0098] Figure 8 illustrates how the hash value is computed. First a median value of each pixel location contributing to the video fingerprint is found by summing the values of said location over a period of many days of collection values at said location from a plurality of television channels representative of the typical television programming to be identified by the invention. Once the median value is determined is can be used indefinitely as a constant without further calculation or adjustment. The pixel value sent from the client to the server matching system is first processed by subtracting the median value of said pixel location. The resulting value is stored with the other pixels locations of a video frame in matrix and an appropriate hashing function is applied to said matrix. Hash values are then derived from the resulting dot product.
[0099] Figure 9 illustrates the beneficial results of using the median values of a pixel location as part of the process of computing the hash values. Chart 901 shows the resulting curve of the output of a typical un-optimized hash function with a relatively small number of hash values occupying a relatively narrow range on the left edge of the curve. The resulting median value 902 is relatively low. Chart 903 shows the favorable redistribution of hash values as a resulting of computing the median of each pixel location that participates in the matching process and applying said median value as part of the hashing function. The distribution of hash values is more spread out with an attendant rise in the median value of all hash keys 904.
[00100] Figure 9a illustrates what happens to a dataset when a median value is not found prior to partitioning said dataset. If the system sampled sixteen pixel locations of each video frame and if each pixel location had a red, green and blue pixel value, there would be 64 dimensions (or axis) to the graph. For the sake of illustration, in this example, the dataset includes just two pixel sample points of a single video frame 906 and 908. Further, the example assumes just a single luminance value is obtained at each pixel point. By splitting the dataset diagonally 907 into clockwise sector 907c and counterclockwise sector 907cc and with vertical 908 and horizontal 906 axis crossing at the zero value 905, there are only two of eight sectors 910 and 911 containing data between the two said pixel locations.
[00101] Figure 9b illustrates the benefit of finding the median value of each pixel location. This example continues to use the assumption that the pixel values are a single luminance value from zero to 255, although absolute value is of no consequence to this method. This example illustrates a simplistic assumption of the median value is 128 for both pixel locations. Now, by shifting the point of partition to 905', the vertical and horizontal axis shift to 908' and 906' respectively. Diagonal slice 909 moves to 909'. It is clear from the illustration that now all eight sectors contain data.
[00102] When partition a dataset in this manner, the computed median is not necessarily in the middle of the dataset nor does it need to be. The desired result is to spread out the data so that when said data is partitioned and assigned to individual servers, the system accessing said servers more uniformly. In contrast, the unoptimized data of Figure 9 would have, if partitioned as illustrated among eight servers, seen only two of the eight servers accessed. In the method illustrated by Figure 9b, with the color values at each pixel location and with an example of 16 pixel locations, the actual computation results in the application of 48 median values computed as a 48 dimension graph. Further, the data can be spliced more than once around each median point of the 48 dimension graph as required to partition said data such that said dataset resulting from said slice can be made to fit within the operational constraints of an individual computer server of the system. In any case, data will be found most of the time on the clockwise and counterclockwise side of each partition slice.
[00103] Figure 10 illustrates an operational flow 1000 representing example operations related to addressing a media database using distance associative hashing. In Figure 10 and in following figures that include various examples of operational flows, discussion and explanation may be provided with respect to the above-described examples of Figures 1 through 9, and/or with respect to other examples and contexts. However, it should be understood that the operational flows may be executed in a number of other environments and contexts, and/or in modified versions of Figures 1 through 9. Also, although the various operational flows are presented in the sequence(s) illustrated, it should be understood that the various operations may be performed in other orders than those which are illustrated, or may be performed concurrently.
[00104] After a start operation, the operational flow 1000 moves to operation 1002. Operation 1002 depicts receiving one or more indications of a sample of a video segment. For example, as shown in and/or described with respect to Figures 1 through 9, the indications may be associated with one or more pixel patches from an ingest system. [00105] Then, operation 1004 depicts determining, for at least one patch of the sample of a video segment including at least one or more pixels of the at least one patch, an algorithmically-derived value of the one or more pixels of each patch. For example, as shown in and/or described with respect to Figures 1 through 9, a mean value of the red pixels in each patch, the green pixels in each patch, and the blue pixels in each patch may be computed.
[00106] Then, operation 1006 depicts subtracting a median point value established for each patch from the mean value for each patch. For example, as shown in and/or described with respect to Figures 1 through 9, a median value of each pixel location contributing to the video fingerprint may be found by summing the values of said location over a period of many days of collection values at said location from a plurality of television channels.
[00107] Then, operation 1008 depicts transforming the values resulting from the subtraction using a function pre-derived to distribute the values evenly. For example, as shown in and/or described with respect to Figures 1 through 9, the values resulting from the subtraction populate a matrix. A dot product of that matrix and a pre-derived static matrix may be computed. The pre-derived static matrix may be determined prior to operational flow 1000 being instantiated, and may be optimized mathematically based on past ingested data such that matrices crossed with it will produce more evenly distributed results than results coming directly from the subtraction operation.
[00108] Then, operation 1010 depicts constructing a hash value from the transformed values. For example, as shown in and/or described with respect to Figures 1 through 9, values capable of holding RGB values are reduced to bit form, such that a hash value may be a string of bits.
[00109] Then, operation 1012 depicts referencing a number of most significant bits of the constructed hash value to determine a database sector. For example, as shown in and/or described with respect to Figures 1 through 9, a number of bits may be predetermined so that the predetermined number of bits of a hash value are used for addressing one or more database sectors.
[00110] Then, operation 1014 depicts storing at least the hash value on the determined database sector. For example, as shown in and/or described with respect to Figures 1 through 9, the hash value may be stored in a bucket, the bucket including other hash values which are mathematically near, where the hash values are associated at least with particular video segments and offsets.
[00111] Figure 11 illustrates alternative embodiments of the example operational flow 1000 of Figure 10. Figure 11 illustrates an example embodiment where operational flow 1000 may include at least one additional operation. Additional operations may include operation 1102.
[00112] Operation 1102 illustrates at least one of the receiving 1002, determining 1004, subtracting 1006, transforming 1008, constructing 1010, referencing 1012, or storing 1014 operations being at least partially implemented using one or more processing devices. In some instances, one of the foregoing operations may be at least partially implemented using one or more computer processors. Other processing devices may include Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs), digital signal processors (DSPs), or any other circuitry configured to effect the result of at least one of the foregoing operations.
[00113] Figure 12 illustrates alternative embodiments of the example operational flow 1000 of Figure 10. Figure 12 illustrates an example embodiment where operation 1002 may include at least one additional operation. Additional operations may include operation 1202, and/or operation 1204.
[00114] Operation 1202 illustrates receiving one or more indications of at least one of a frame or a still image. For example, as shown in and/or described with respect to Figures 1 through 9, a sample of a video segment may be comprised of an individual frame of a video stream. Such a frame may be one 30fps video frame. In different embodiments, a sample of a video segment may be a still image, or a portion of a video segment that may be imaged at a rate other than 30 times a second.
[00115] Further, operation 1204 illustrates receiving one or more indications of a sample of a video segment, the one or more indications of a sample of a video segment associated with at least one indication of a channel, at least one indication of a video segment, and at least one indication of a timecode offset from the beginning of the video segment. For example, as shown in and/or described with respect to Figures 1 through 9, data associated with a video segment (which may be a program title and/or other metadata associated with a video segment), the channel from which the program was ingested, and an offset in time from the start of the program may be received, from, for example, a channel guide associated with a channel which is being monitored by the ingest system.
[00116] Figure 13 illustrates alternative embodiments of the example operational flow 1000 of Figure 10. Figure 13 illustrates an example embodiment where operation 1004 may include at least one additional operation 1302.
[00117] Operation 1302 illustrates determining, for at least one patch of the sample of a video segment including at least one or more pixels of the at least one patch, a mean value of the one or more pixels of each patch. For example, as shown in and/or described with respect to Figures 1 through 9, the algorithmic operation used to reduce the one or more pixels in a patch to a single value may be, for example, an arithmetic mean.
[00118] Figure 14 illustrates alternative embodiments of the example operational flow 1000 of Figure 10. Figure 14 illustrates an example embodiment where operation 1006 may include at least one additional operation 1402.
[00119] Operation 1402 illustrates subtracting a median point value established for each patch from the mean value for each patch, the median point value established for each patch having been previously determined utilizing data from each patch for a plurality of channels over at least one period of time. For example, as shown in and/or described with respect to Figures 1 through 9, a median value may be determined, the median value determined for each patch, wherein medians are established for the same patches at ingest as in the operation of determining a segment on a client system, the median being established as a constant value derived from monitoring the same patches across many channels for a long time (a month, a year, etc.).
[00120] Figure 15 illustrates alternative embodiments of the example operational flow 1000 of Figure 10. Figure 15 illustrates an example embodiment where operation 1008 may include at least one additional operation. Additional operations may include operation 1502, operation 1504, and/or operation 1506.
[00121] Operation 1502 illustrates forming a variable matrix including at least the values resulting from the subtraction. For example, as shown in and/or described with respect to Figures 1 through 9, values are arranged in a matrix, the values resulting from the subtraction operation, wherein the subtraction operation subtracts the median value established over time for each patch from the mean value of the instant frame being ingested.
[00122] Operation 1504 illustrates obtaining a static matrix which, when crossed with the variable matrix, will more evenly distribute the transformed values. For example, as shown in and/or described with respect to Figures 1 through 9, a matrix may be determined based upon mathematical analysis of previously-obtained data sets related to hash values. The matrix may be optimized mathematically such that, when used as an operand in a dot product operation with successive variable matrices, the corresponding successive result matrices will include values that are more evenly spread along a distribution curve than the variable matrices prior to the dot product operation.
[00123] Operation 1506 illustrates computing a dot product of the variable matrix and the static matrix, the dot product including at least the more evenly-distributed transformed values. For example, as shown in and/or described with respect to Figures 1 through 9, the variable matrix containing values resulting from the subtraction operation may be crossed with a static matrix that has been predetermined to distribute data represented by a variable matrix more evenly, such that the resulting matrices are more spread out instead of being bunched about a particular portion of the distribution.
[00124] Figure 16 illustrates alternative embodiments of the example operational flow 1000 of Figure 10. Figure 16 illustrates an example embodiment where operation 1504 may include at least one additional operation 1602.
[00125] Operation 1602 illustrates determining, using locality-sensitive hashing at least partially based on one or more previously obtained hash values, a static matrix which, when crossed with a variable matrix, will more evenly distribute the transformed values of the variable matrix. For example, as shown in and/or described with respect to Figures 1 through 9, a locality-sensitive hashing technique may be used to analyze previously-ingested video samples, producing a matrix such that, when used as an operand in a dot product operation with successive variable matrices, the corresponding successive result matrices will include values that are more evenly spread along a distribution curve than the variable matrices prior to the dot product operation. [00126] Figure 17 illustrates alternative embodiments of the example operational flow 1000 of Figure 10. Figure 17 illustrates an example embodiment where operation 1010 may include at least one additional operation. Additional operations may include operation 1702, and/or operation 1704.
[00127] Operation 1702 illustrates constructing a hash value from the transformed values, including at least reducing the fidelity of the transformed values via reducing each transformed value to a binary representation. For example, as shown in and/or described with respect to Figures 1 through 9, each value of the resultant matrix from the dot product operation may be reduced from, for example, an 8-bit value from 0 to 255 (or from -127 to 128) to a single bit, being either a one or a zero.
[00128] Operation 1702 may include operation 1704. Operation 1704 illustrates determining for each transformed value whether the transformed value is a positive number and, if the transformed value is a positive number, assigning a one to the hash value and otherwise assigning a zero to the hash value. For example, as shown in and/or described with respect to Figures 1 through 9, each value of the resultant matrix from the dot product operation between 1 and 128 may be reduced to a bit value of 1, and each value of the resultant matrix from the dot product operation between -127 and 0 may be reduced to a bit value of 0.
[00129] Figure 18 illustrates alternative embodiments of the example operational flow 1000 of Figure 10. Figure 18 illustrates an example embodiment where operation 1012 may include at least one additional operation 1802.
[00130] Operation 1802 illustrates referencing a number of most significant bits of the constructed hash value to determine a database server, wherein the number of most significant bits is pre-determined to address a plurality of database servers, wherein a number of database servers associated with the number of most significant bits is established to enable at least one index associated with a database sector to reside entirely in memory of a corresponding database server. For example, as shown in and/or described with respect to Figures 1 through 9, a number of most significant bits of 2 may be selected, whereby the 2 bits may provide four different values (00, 01, 10, and 11), each of which may be assigned to a different database sector. The number of most significant bits of a hash value may be established to provide a sufficient number of servers such that a content associated with a plurality of hash values may fit entirely in the memory of a particular database sector, which may be a database server, a cluster partner, a virtual machine, and/or another type of database node. The number of bits does not have to, but may, exactly represent the maximum number of database sectors at any given time (i.e. while 6 bits may be selected to provide for addressing of up to 64 database sectors, the system may be operable with fewer servers e.g. 60 sectors, or with the maximum 64 sectors).
[00131] Figure 19 illustrates alternative embodiments of the example operational flow 1000 of Figure 10. Figure 19 illustrates an example embodiment where operation 1014 may include at least one additional operation 1902.
[00132] Operation 1902 illustrates storing at least the hash value on the determined database sector, including at least storing at least one indication of a channel, at least one indication of a video segment, and at least one indication of a timecode offset from the beginning of the video segment at a database location at least partially based on the hash value. For example, as shown in and/or described with respect to Figures 1 through 9, data associated with a video segment (which may be a program title and/or other metadata associated with a video segment), the channel from which the program was ingested, and an offset in time from the start of the program may be stored, either along with the hash value or in a location associated with and/or referenceable by the hash value, the storage being in the same or different sector, server, or database as the hash value.
[00133] Figure 20 illustrates an operational flow 2000 representing example operations related to addressing a media database using distance associative hashing. In Figure 20 and in following figures that include various examples of operational flows, discussion and explanation may be provided with respect to the above-described examples of Figures 1 through 9, and/or with respect to other examples and contexts. However, it should be understood that the operational flows may be executed in a number of other environments and contexts, and/or in modified versions of Figures 1 through 9. Also, although the various operational flows are presented in the sequence(s) illustrated, it should be understood that the various operations may be performed in other orders than those which are illustrated, or may be performed concurrently.
[00134] After a start operation, the operational flow 2000 moves to operation 2002. Operation 2002 depicts receiving a cue, the cue constructed via one or more operations associated with a media storage operation. For example, as shown in and/or described with respect to Figures 1 through 9, at least some data is received which is associated with a sample of video data taken from a particular client system. The data may be associated with exactly the same patches of the client system as are defined by the ingest operation. The data may be algorithmically processed to arrive at a hash value using the same operations as the ingest operation. Accordingly, if a particular frame associated with a particular time offset of a particular program on a particular channel is ingested and hashed, resulting in a hash value associated with that particular frame, should that particular frame also be sampled while being displayed on a client system, the same hashing operations as applied to the ingested frame will result in the same hash value as resulted from the hashing operations on the ingested frame. But in contrast to the hash value prepared during the ingest, the cue of operation 2002 represents data associated with a sample of video data from a particular client system. A cue may be received via, for example, an HTTP request.
[00135] Then, operation 2004 depicts referencing a number of most significant bits of the received cue to determine a database sector. For example, as shown in and/or described with respect to Figures 1 through 9, the same bits of the cue are examined as defined by the number of most significant bits used to reference a database sector during ingest. For example, if the first two bits of the hash value at ingest are used for storing the hash value at a particular database sector, the same first two bits of the cue associated with a sample of video data from a client system are used for addressing a particular database sector.
[00136] Then, operation 2006 depicts returning at least one indication of at least one candidate from the database sector based at least partially on the received cue. For example, as shown in and/or described with respect to Figures 1 through 9, hash values which exactly match the cue, or are nearby the cue, are returned as one or more of suspects or candidates. Candidates may be returned within a particular percentage radius. Candidates may be returned according to a nearest neighbor algorithm or a modified nearest neighbor algorithm.
[00137] Figure 21 illustrates alternative embodiments of the example operational flow 2000 of Figure 20. Figure 21 illustrates an example embodiment where operation 2002 may include at least one additional operation. Additional operations may include operation 2102, operation 2104, and/or operation 2106. [00138] Operation 2102 illustrates receiving a cue associated with a sample of a video buffer of a client system, including at least receiving one or more indications related to an epoch time associated with the sample of the video buffer of the client system. For example, as shown in and/or described with respect to Figures 1 through 9, a cue may include, or be associated with, a time offset from an arbitrary time. The time offset may be computed from January 1, 1970, for example.
[00139] Operation 2104 illustrates receiving a cue, the cue associated with a sample of a video buffer of a client system, the cue at least partially determined by hashing at least some values associated with the video buffer. For example, as shown in and/or described with respect to Figures 1 through 9, patches associated with a video buffer may be reduced to a bit string via one or more mathematical operations or algorithms using one or more operands as constants, the constants pre-derived via operations described elsewhere herein with respect to hashing, for example.
[00140] Operation 2106 illustrates receiving a cue, the cue associated with a sample of a video buffer of a client system, the cue at least partially determined by hashing at least some values associated with the video buffer, the hashing based at least partially one or more of at least one operand or at least one algorithm also utilized in an associated media storage operation. For example, as shown in and/or described with respect to Figures 1 through 9, at least some data associated with a sample of a video buffer representing what is displayed by a television screen at a particular quantum of time is processed via operations utilized by the ingest process and/or in conjunction with data locations common to the ingest process and/or involving constant values for operands utilized by the ingest process. For example, the number of patches analyzed at ingest may also be utilized in providing a cue associated with a particular client system. The size of pixel patches analyzed at ingest may also be utilized in providing a cue associated with a particular client system. The same pre-derived static matrix used to more evenly distribute hash values at ingest may also be used during hashing of the data associated with a particular client system.
[00141] Figure 22 illustrates alternative embodiments of the example operational flow 2000 of Figure 20. Figure 22 illustrates an example embodiment where operation 2002 may include at least one additional operation. Additional operations may include operation 2202, operation 2204, operation 2206, operation 2208, operation 2210, operation 2212, and/or operation 2214.
[00142] Operation 2202 illustrates receiving one or more indications of at least one content of a video buffer of a client system. For example, as shown in and/or described with respect to Figures 1 through 9, pixel values for red, green, and blue pixels at every pixel location at every pre-defined patch of the video buffer of the client system may be read, for every frame, or for every third frame, or for every tenth frame, or for every second, or at some other interval. The indications (pixel values or other data) may be received by a widget on the television, by control logic on the television, by a system coupled with the media server, or elsewhere.
[00143] Operation 2204 illustrates determining, for at least one patch of the at least one content of the video buffer including at least one or more pixels of the at least one patch, an algorithmically-derived value of the one or more pixels of each patch. For example, as shown in and/or described with respect to Figures 1 through 9, pixel values for red, green, and blue pixels at every pixel location at every pre-defined patch of the video buffer of the client system may be averaged.
[00144] Operation 2206 illustrates subtracting a median point value from the mean value for each patch. For example, as shown in and/or described with respect to Figures 1 through 9, median point values at each patch established through analysis of ingested content are determined. The median point values for each patch may, for example, be provided to the client system once determined by a system associated with the media database and ingest system. The median point values may be updated from time to time (hourly, daily, monthly, yearly). The median point values provided for hashing data associated with a video buffer of a client system may be the same median point values utilized to hash incoming content at ingest.
[00145] Operation 2208 illustrates transforming the values resulting from the subtraction. For example, as shown in and/or described with respect to Figures 1 through 9, values resulting from the subtraction are populated in a matrix and crossed with a pre-defined static matrix. The dot-product operation crossing the two matrices may be conducted at the client system during a process of converting pixel patch data associated with a frame in a video buffer to a cue, such that a cue is sent in an HTTP request rather than the actual pixel patch data, resulting in a compact HTTP message. The pre-defined static matrix may be provided to the client system in advance of the transform, and may be the same matrix as was produced to distribute hashed values at ingest more evenly. The pre-defined static matrix may be updated at the client system from time to time. Alternatively, patch data may be sent, with or without other metadata, from a client system (television, e.g.) to a different system for processing and/or hashing.
[00146] Operation 2210 illustrates constructing a hash value from the transformed values. For example, as shown in and/or described with respect to Figures 1 through 9, the values in the matrix resulting from crossing the matrix with values associated with the video buffer with the pre-derived static matrix may be reduced to bits, with a single bit replacing each 8-bit value in the matrix. In other embodiments, the constructed hash value may include a different number of bits for each value in the matrix. In different embodiments, the constructed hash value may have the same number of bits as the values in the matrix, or may be a direct representation of the values in the matrix.
[00147] Operation 2212 illustrates associating the cue at least partially with the constructed hash value. For example, as shown in and/or described with respect to Figures 1 through 9, the string of bits constructed from the transformed matrix may be a cue, or may associate the constructed string of bits with a time (such as an epoch time) to form a cue, or may associate other data such as an IP address or other identifier associated with the client television or a widget of the client television to form a cue. Alternatively, the cue may include or otherwise be associated with any other metadata associated with audiovisual content at the client system.
[00148] Operation 2214 illustrates at least one of the determining 2204, subtracting 2206, transforming 2208, or constructing 2210 operations utilize one or more of at least one operand or at least one algorithm also utilized in an associated media storage operation. For example, as shown in and/or described with respect to Figures 1 through 9, one or more parameters including one or more of a definition of a number of pixel patches, a definition of a size of pixel patches, a pre-defined median value associated with pixel patches, or a pre-defined static matrix may be provided to a client TV, the one or more parameters also utilized by the ingest process such that operations applied to a sample from a video buffer will result in the same hash value that would result when that frame (e.g. same video segment and time offset) was ingested and hashed. [00149] Figure 23 illustrates alternative embodiments of the example operational flow 2000 of Figure 20. Figure 23 illustrates an example embodiment where operation 2006 may include at least one additional operation. Additional operations may include operation 2302, and/or operation 2304.
[00150] Operation 2302 illustrates returning at least one indication of at least one candidate from the database sector based at least partially on a probabilistic point location in equal balls ("PPLEB") algorithm as a function of the received cue. For example, as shown in and/or described with respect to Figures 1 through 9, at least one of candidates or suspects representing path points close to a cue (e.g. neighbors, nearest neighbors, within a radius, from within the same bucket, belonging to the same ring, etc.) are returned from a media database constructed and/or modified via an ingest process.
[00151] Operation 2304 illustrates returning at least one indication of at least one candidate from the database sector based at least partially on the received cue, the at least one candidate being within a predetermined inverse percentage distribution radius of the received cue. For example, as shown in and/or described with respect to Figures 1 through 9, at least one of candidates or suspects associated with locality sensitive hashing related to at least one of a cue or a hash value are returned.
[00152] Figure 24 illustrates an operational flow 2400 representing example operations related to addressing a media database using distance associative hashing. In Figure 24 and in following figures that include various examples of operational flows, discussion and explanation may be provided with respect to the above-described examples of Figures 1 through 9, and/or with respect to other examples and contexts. However, it should be understood that the operational flows may be executed in a number of other environments and contexts, and/or in modified versions of Figures 1 through 9. Also, although the various operational flows are presented in the sequence(s) illustrated, it should be understood that the various operations may be performed in other orders than those which are illustrated, or may be performed concurrently.
[00153] After a start operation, the operational flow 2400 moves to operation 2402. Operation 2402 depicts receiving at least one indication of at least one candidate and at least one indication of at least one cue. For example, as shown in and/or described with respect to Figures 1 through 9, a hash value related to a video buffer of a client system, along with one or more associated candidates or suspects is determined.
[00154] Then, operation 2404 depicts adding a token to a bin associated with at least one received candidate. For example, as shown in and/or described with respect to Figures 1 through 9, scoring of candidates is performed via tokens added to bins corresponding to candidates/suspects, the token being, for example, a value which is incremented each time a token is added.
[00155] Then, operation 2406 depicts determining whether a number of tokens in a bin exceeds a value associated with a probability that a client system is displaying a particular video segment associated with at least one cue and, if the number of tokens in a bin exceeds a value associated with a probability that a client system is display a particular video segment associated with at least one cue, returning at least some data associated with the particular video segment based at least partially on the bin. For example, as shown in and/or described with respect to Figures 1 through 9, a determination of a particular video segment and particular offset of the video segment is probabilistically determined via the scoring associated with the bins.
[00156] Figure 25 illustrates alternative embodiments of the example operational flow 2400 of Figure 24. Figure 25 illustrates an example embodiment where operation 2404 may include at least one additional operation 2502.
[00157] Operation 2502 illustrates adding a token to a time bin associated with at least one received candidate. For example, as shown in and/or described with respect to Figures 1 through 9, a data structure associated with a candidate/suspect may include an arbitrary time bin grouped by an arbitrary time.
[00158] Figure 26 illustrates alternative embodiments of the example operational flow 2400 of Figure 20. Figure 26 illustrates an example embodiment where operation 2404 may include at least one additional operation. Additional operations may include operation 2602, and/or operation 2604. Further, operational flow 2400 may include at least one additional operation 2606.
[00159] Operation 2602 illustrates determining a relative time, including at least subtracting a candidate time associated with the at least one candidate from an arbitrary time associated with the at least one cue. For example, as shown in and/or described with respect to Figures 1 through 9, a time offset of a video segment associated with a candidate is subtracted from an arbitrary time associated with an epoch time related to the cue received from a client system (television, set-top box, or article, machine, or composition of matter displaying and/or providing and/or receiving video content).
[00160] Operation 2604 illustrates adding a token to a time bin associated with the candidate based at least partially on the determined relative time. For example, as shown in and/or described with respect to Figures 1 through 9, when a cue point associated with the client system matches or nearly matches a reference cue point associated with a media database, a token may be added to a bin, which may include incrementing a value associated with a bin or another means of tracking bin operations.
[00161] Operation 2606 illustrates removing one or more tokens from a time bin based at least partially on a time period elapsing. For example, as shown in and/or described with respect to Figures 1 through 9, a bin may be leaky such that data and/or tokens associated with old suspects/candidates may be release from the bin, which may include decrementing a value associated with a bin or another means of tracking bin operations.
[00162] In varying embodiments, pixel locations may relate to one or many colors and/or color spaces/models (e.g. red, blue, green; red, blue, green, and yellow; cyan, magenta, yellow, and black; a single pixel value uniquely identifying a color e.g. a 24-bit value associated with a pixel location; hue, saturation, brightness; etc.). Differing numbers of pixels in a patch may be used, and the patch does not have to be a square patch. Further, resolution of the video buffer of the client system may vary. Resolutions and/or color densities at the client system and the ingest system may vary. The system may be operable with various raster resolutions, including but not limited to 1920 by 1080, 3840 by 2160, 1440 x 1080, 1366 x 768, or other resolutions. It is expected that over the next two decades, increases in pixel resolution of common programming, televisions, and/or client systems will occur; the same basic operations may be utilized although pixel patch number, size, sampling rate, or other aspects may vary. Further, an up-conversion, down-conversion, or other transformation operation associated with resolution and/or color density may occur and/or be interposed between other operations described herein. [00163] Figure 27 illustrates an example system 2700 in which embodiments may be implemented. The system 2700 includes one or more computing devices 2702. The system 2700 also illustrates a fabric 2704 for facilitating communications among one or more computing devices and one or more client devices 2706. The system 2700 also illustrates one or more client devices 2706. In some embodiments, the one or more client devices may be among the one or more computing devices. The system 2700 also illustrates at least one non-transitory computer- readable medium 2708. In some embodiments, 2708 may include one or more instructions 2710 that, when executed on at least some of the one or more computing devices, cause at least some of the one or more computing devices to at least receive at least one stream of rasterized video; create at least one hash value associated with at least one sample of at least one received rasterized video stream; determine at least one database sector for storing a created at least one hash value; and store a created at least one hash value on at least one determined database sector. In differing embodiments, the one or more instructions may be executed on a single computing device. In other embodiments, some portions of the one or more instructions may be executed by a first plurality of the one or more computing devices, while other portions of the one or more instructions may be executed by a second plurality of the one or more computing devices.
[00164] Figure 28 illustrates an example system 2800 in which embodiments may be implemented. The system 2800 includes one or more computing devices 2802. The system 2800 also illustrates a fabric 2804 for facilitating communications among one or more computing devices and one or more client devices 2806. The system 2800 also illustrates one or more client devices 2806. In some embodiments, the one or more client devices may be among the one or more computing devices. The system 2800 also illustrates at least one non-transitory computer- readable medium 2808. In some embodiments, 2808 may include one or more instructions 2810 that, when executed on at least some of the one or more computing devices, cause at least some of the one or more computing devices to at least receive one or more indications associated with at least one video buffer of at least one client system; determine a cue based at least partially on the at least one video buffer and at least one epoch time associated with the at least one video buffer, wherein one or more of at least one operand or at least one function associated with determining the cue is also utilized in an associated media storage operation; reference a number of most significant bits of a determined cue to determine a database sector; and return at least one indication of at least one candidate from a determined database sector based at least partially on a determined cue. In differing embodiments, the one or more instructions may be executed on a single computing device. In other embodiments, some portions of the one or more instructions may be executed by a first plurality of the one or more computing devices, while other portions of the one or more instructions may be executed by a second plurality of the one or more computing devices.
[00165] Figure 29 illustrates an example system 2900 in which embodiments may be implemented. The system 2900 includes one or more computing devices 2902. The system 2900 also illustrates a fabric 2904 for facilitating communications among one or more computing devices and one or more client devices 2906. The system 2900 also illustrates one or more client devices 2906. In some embodiments, the one or more client devices may be among the one or more computing devices. The system 2900 also illustrates at least one non-transitory computer- readable medium 2908. In some embodiments, 2908 may include one or more instructions 2910 that, when executed on at least some of the one or more computing devices, cause at least some of the one or more computing devices to at least receive at least one indication of at least one candidate and at least one indication of at least one cue; add a token to a bin associated with at least one received candidate; and determine whether a number of tokens in a bin exceeds a value associated with a probability that a client system is receiving a particular video segment associated with at least one received cue and, if the number of tokens in a bin exceeds a value associated with a probability that a client system is receiving a particular video segment associated with at least one received cue, returning at least some data associated with the particular video segment based at least partially on the bin. In differing embodiments, the one or more instructions may be executed on a single computing device. In other embodiments, some portions of the one or more instructions may be executed by a first plurality of the one or more computing devices, while other portions of the one or more instructions may be executed by a second plurality of the one or more computing devices.
[00166] Certain aspects of the present invention include process steps and instructions described herein in the form of an algorithm. It should be noted that the process steps and instructions of the present invention could be embodied in software, firmware or hardware, and when embodied in software, could be downloaded to reside on and be operated from different platforms used by real-time network operating systems.
[00167] The present invention also relates to an apparatus for performing the operations herein. This apparatus may be specially constructed for the required purposes, or it may comprise a general-purpose computer selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored in a computer readable storage medium, such as, but is not limited to, any type of disk including floppy disks, optical disks, CD-ROMs, magnetic-optical disks, read-only memories (ROMs), random access memories (RAMs), EPROMs, EEPROMs, magnetic or optical cards, application specific integrated circuits (ASICs), or any type of media suitable for storing electronic instructions, and each coupled to a computer system bus.
[00168] Furthermore, computers or computing means referred to in the specification may include a single processor or may employ multiple-processor designs for increased computing capability.
[00169] The algorithms and displays presented herein are not inherently related to any particular computer or other apparatus. Various general-purpose systems may also be used with programs in accordance with the teachings herein, or it may prove convenient to construct more specialized apparatus to perform the required method steps. The required structure for a variety of these systems will appear from the description above. In addition, the present invention is not described with reference to any particular programming language or operating systems. It is appreciated that a variety of programing languages and operating systems may be used to implement the teachings of the present invention as described herein.
[00170] The system and methods, flow diagrams, and structure block diagrams described in this specification may be implemented in computer processing systems including program code comprising program instructions that are executable by a computer processing system. Other implementations may also be used. Additionally, the flow diagrams and structure block diagrams herein described describe particular methods and/or corresponding acts in support of steps and corresponding functions in support of disclosed structural means, may also be utilized to implement corresponding software structures and algorithms, and equivalents thereof. [00171] Embodiments of the subject matter described in this specification can be implemented as one or more computer program products, i.e., one or more modules of computer program instructions encoded on a tangible program carrier for execution by, or to control the operation of, data processing apparatus. The computer readable medium can be a machine readable storage device, a machine readable storage substrate, a memory device, or a combination of one or more of them.
[00172] A computer program (also known as a program, software, software application, script, or code) can be written in any form of programming language, including compiled or interpreted languages, or declarative or procedural languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program does not necessarily correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a suitable communication network.
[00173] The processes and logic flows described in this specification can be performed by one or more programmable processors executing one or more computer programs to perform functions by operating on input data and generating output. The processes and logic flows can also be performed by, and apparatus can also be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit).
[00174] The essential elements of a computer are a processor for performing instructions and one or more memory devices for storing instructions and data. Generally, a computer will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magneto optical disks, or optical disks. However, a computer need not have such devices. Processors suitable for the execution of a computer program include, by way of example only and without limitation, both general and special purpose microprocessors, and any one or more processors of any kind of digital computer. Generally, a processor will receive instructions and data from a read only memory or a random access memory or both.
[00175] To provide for interaction with a user or manager of the system described herein, embodiments of the subject matter described in this specification can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor, for displaying information to the user and a keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well. For example, feedback provided to the user can be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including acoustic, speech, or tactile input.
[00176] Embodiments of the subject matter described in this specification can be implemented in a computing system that includes back end component(s) including one or more data servers, or that includes one or more middleware components such as application servers, or that includes a front end component such as a client computer having a graphical user interface or a Web browser through which a user or administrator can interact with some implementations of the subject matter described is this specification, or any combination of one or more such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication, such as a communication network. The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client server relationship to each other.
[00177] While this specification contains many specific implementation details, these should not be construed as limitations on the scope of any invention or of what may be claimed, but rather as descriptions of features that may be specific to particular embodiments of particular inventions. Certain features that are described in this specification in the context of separate embodiments can also be implemented in combination in a single embodiment.
[00178] Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination. Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a subcombination or variation of a subcombination.
[00179] Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Moreover, the separation of various system components in the embodiments described above should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.
[00180] This written description sets forth the best mode of the invention and provides examples to describe the invention and to enable a person of ordinary skill in the art to make and use the invention. This written description does not limit the invention to the precise terms set forth. Thus, while the invention has been described in detail with reference to the examples set forth above, those of ordinary skill in the art may effect alterations, modifications and variations to the examples without departing from the scope of the invention.

Claims

What is Claimed is:
1. A method, comprising:
receiving one or more indications of a sample of a video segment;
determining, for at least one patch of the sample of a video segment including at least one or more pixels of the at least one patch, an algorithmically-derived value of the one or more pixels of each patch;
subtracting a median point value established for each patch from the mean value for each patch;
transforming the values resulting from the subtraction using a function pre-derived to distribute the values evenly;
constructing a hash value from the transformed values;
referencing a number of most significant bits of the constructed hash value to determine a database sector; and
storing at least the hash value on the determined database sector.
2. The method of claim 1, wherein at least one of the receiving, determining, subtracting, transforming, constructing, referencing, or storing is at least partially implemented using one or more processing devices.
3. The method of claim 1, wherein receiving one or more indications of a sample of a video segment comprises:
receiving one or more indications of at least one of a frame or a still image.
4. The method of claim 1, wherein receiving one or more indications of a sample of a video segment comprises:
receiving one or more indications of a sample of a video segment, the one or more
indications of a sample of a video segment associated with at least one indication of a channel, at least one indication of a video segment, and at least one indication of a timecode offset from the beginning of the video segment.
5. The method of claim 1, wherein determining, for at least one patch of the sample of a video segment including at least one or more pixels of the at least one patch, an algorithmically-derived value of the one or more pixels of each patch comprises:
determining, for at least one patch of the sample of a video segment including at least one or more pixels of the at least one patch, a mean value of the one or more pixels of each patch.
6. The method of claim 1, wherein subtracting a median point value established for each patch from the mean value for each patch comprises:
subtracting a median point value established for each patch from the mean value for each patch, the median point value established for each patch having been previously determined utilizing data from each patch for a plurality of channels over at least one period of time.
7. The method of claim 1, wherein transforming the values resulting from the subtraction using a function pre-derived to distribute the values evenly comprises:
forming a variable matrix including at least the values resulting from the subtraction; obtaining a static matrix which, when crossed with the variable matrix, will more evenly distribute the transformed values; and
computing a dot product of the variable matrix and the static matrix, the dot product including at least the more evenly-distributed transformed values.
8. The method of claim 7, wherein obtaining a static matrix which, when crossed with the variable matrix, will more evenly distribute the transformed values comprises:
determining, using locality-sensitive hashing at least partially based on one or more
previously obtained hash values, a static matrix which, when crossed with a variable matrix, will more evenly distribute the transformed values of the variable matrix.
9. The method of claim 1, wherein constructing a hash value from the transformed values comprises:
constructing a hash value from the transformed values, including at least reducing the fidelity of the transformed values via reducing each transformed value to a binary representation.
10. The method of claim 9, wherein reducing the fidelity of the transformed values via reducing each transformed value to a binary representation comprises:
determining for each transformed value whether the transformed value is a positive
number and, if the transformed value is a positive number, assigning a one to the hash value and otherwise assigning a zero to the hash value.
11. The method of claim 1 , wherein referencing a number of most significant bits of the constructed hash value to determine a database sector comprises:
referencing a number of most significant bits of the constructed hash value to determine a database server, wherein the number of most significant bits is pre-determined to address a plurality of database servers, wherein a number of database servers associated with the number of most significant bits is established to enable at least one index associated with a database sector to reside entirely in memory of a corresponding database server.
12. The method of claim 1, wherein storing at least the hash value on the determined database sector comprises:
storing at least the hash value on the determined database sector, including at least storing at least one indication of a channel, at least one indication of a video segment, and at least one indication of a timecode offset from the beginning of the video segment at a database location at least partially based on the hash value.
13. A method, comprising :
receiving a cue, the cue constructed via one or more operations associated with a media storage operation;
referencing a number of most significant bits of the received cue to determine a database sector; and
returning at least one indication of at least one candidate from the database sector based at least partially on the received cue.
14. The method of claim 13, wherein receiving a cue, the cue constructed via one or more operations associated with a media storage operation comprises:
receiving a cue associated with a sample of a video buffer of a client system, including at least receiving one or more indications related to an epoch time associated with the sample of the video buffer of the client system.
15. The method of claim 13, wherein receiving a cue, the cue constructed via one or more operations associated with a media storage operation comprises:
receiving a cue, the cue associated with a sample of a video buffer of a client system, the cue at least partially determined by hashing at least some values associated with the video buffer.
16. The method of claim 15, wherein receiving a cue, the cue associated with a sample of a video buffer of a client system, the cue at least partially determined by hashing at least some values associated with the video buffer comprises:
receiving a cue, the cue associated with a sample of a video buffer of a client system, the cue at least partially determined by hashing at least some values associated with the video buffer, the hashing based at least partially one or more of at least one operand or at least one algorithm also utilized in an associated media storage operation.
17. The method of claim 13, wherein receiving a cue, the cue constructed via one or more operations associated with a media storage operation comprises:
receiving a cue, the cue determined via one or more operations including at least:
receiving one or more indications of at least one content of a video buffer of a client system;
determining, for at least one patch of the at least one content of the video buffer including at least one or more pixels of the at least one patch, an algorithmically-derived value of the one or more pixels of each patch;
subtracting a median point value from the mean value for each patch; transforming the values resulting from the subtraction;
constructing a hash value from the transformed values; and
associating the cue at least partially with the constructed hash value, wherein at least one of the determining, subtracting, transforming, or constructing operations utilize one or more of at least one operand or at least one algorithm also utilized in an associated media storage operation.
18. The method of claim 13, wherein returning at least one indication of at least one candidate from the database sector based at least partially on the received cue comprises:
returning at least one indication of at least one candidate from the database sector based at least partially on a probabilistic point location in equal balls ("PPLEB") algorithm as a function of the received cue.
19. The method of claim 13, wherein returning at least one indication of at least one candidate from the database sector based at least partially on the received cue comprises:
returning at least one indication of at least one candidate from the database sector based at least partially on the received cue, the at least one candidate being within a predetermined inverse percentage distribution radius of the received cue.
20. A method, comprising:
receiving at least one indication of at least one candidate and at least one indication of at least one cue;
adding a token to a bin associated with at least one received candidate; and
determining whether a number of tokens in a bin exceeds a value associated with a
probability that a client system is displaying a particular video segment associated with at least one cue and, if the number of tokens in a bin exceeds a value associated with a probability that a client system is display a particular video segment associated with at least one cue, returning at least some data associated with the particular video segment based at least partially on the bin.
21. The method of claim 20, wherein adding a token to a bin associated with at least one received candidate comprises:
adding a token to a time bin associated with at least one received candidate.
22. The method of claim 20, wherein adding a token to a bin associated with at least one received candidate comprises:
determining a relative time, including at least subtracting a candidate time associated with the at least one candidate from an arbitrary time associated with the at least one cue; and
adding a token to a time bin associated with the candidate based at least partially on the determined relative time.
23. The method of claim 21, further comprising:
removing one or more tokens from a time bin based at least partially on a time period elapsing.
24. A system, comprising:
one or more computing devices; and more instructions that, when executed on at least some of the one or more computing devices, cause at least some of the one or more computing devices to at least:
receive at least one stream of rasterized video;
create at least one hash value associated with at least one sample of at least one received rasterized video stream;
determine at least one database sector for storing a created at least one hash value; and
store a created at least one hash value on at least one determined database sector.
PCT/US2014/030782 2013-03-15 2014-03-17 Systems and methods for addressing a media database using distance associative hashing WO2014145929A1 (en)

Priority Applications (23)

Application Number Priority Date Filing Date Title
CA3173549A CA3173549A1 (en) 2013-03-15 2014-03-17 Systems and methods for identifying video segments for displaying contextually relevant content
CN201480017043.9A CN105144141B (en) 2013-03-15 2014-03-17 For using the system and method apart from relevance hashing to media database addressing
CA2906199A CA2906199C (en) 2013-03-15 2014-03-17 Systems and methods for addressing a media database using distance associative hashing
MX2015012512A MX365827B (en) 2013-03-15 2014-03-17 Systems and methods for identifying video segments for displaying contextually relevant content.
CA2906192A CA2906192C (en) 2013-03-15 2014-03-17 Systems and methods for real-time television ad detection using an automated content recognition database
CN201480015936.XA CN105052161B (en) 2013-03-15 2014-03-17 The system and method for real-time television purposes of commercial detection
PCT/US2014/030795 WO2014145938A1 (en) 2013-03-15 2014-03-17 Systems and methods for real-time television ad detection using an automated content recognition database
BR112015023380-5A BR112015023380B1 (en) 2013-03-15 2014-03-17 SYSTEM AND METHOD FOR REAL-TIME TELEVISION ADVERTISING DETECTION USING AUTOMATED CONTENT RECOGNITION DATABASE
CN201811395356.4A CN110083739B (en) 2013-03-15 2014-03-17 System and method for addressing media databases using distance associative hashing
BR112015023369-4A BR112015023369B1 (en) 2013-03-15 2014-03-17 Computer-implemented system and method
PCT/US2014/030805 WO2014145947A1 (en) 2013-03-15 2014-03-17 Systems and methods for identifying video segments for displaying contextually relevant content
CA2906173A CA2906173C (en) 2013-03-15 2014-03-17 Systems and methods for identifying video segments for displaying contextually relevant content
MX2015012511A MX366327B (en) 2013-03-15 2014-03-17 Systems and methods for real-time television ad detection using an automated content recognition database.
BR112015023389-9A BR112015023389B1 (en) 2013-03-15 2014-03-17 METHOD AND SYSTEM FOR IDENTIFYING VIDEO SEGMENTS TO DISPLAY CONTEXTUALLY RELEVANT CONTENT
MX2015012510A MX356884B (en) 2013-03-15 2014-03-17 Systems and methods for addressing a media database using distance associative hashing.
MX2020001441A MX2020001441A (en) 2013-03-15 2014-03-17 Systems and methods for addressing a media database using distance associative hashing.
EP14762850.7A EP3001871B1 (en) 2013-03-15 2014-03-17 Systems and methods for addressing a media database using distance associative hashing
MX2019008020A MX2019008020A (en) 2013-03-15 2015-09-11 Systems and methods for real-time television ad detection using an automated content recognition database.
CL2015002623A CL2015002623A1 (en) 2013-03-15 2015-09-11 Systems and methods to identify video segments to visualize contextually relevant content
CL2015002619A CL2015002619A1 (en) 2013-03-15 2015-09-11 Systems and methods for detecting real-time television ads using an automated content recognition database.
MX2019007031A MX2019007031A (en) 2013-03-15 2015-09-11 Systems and methods for identifying video segments for displaying contextually relevant content.
HK16105168.7A HK1218193A1 (en) 2013-03-15 2016-05-05 Systems and methods for real-time television ad detection using an automated content recognition database
HK16105782.3A HK1217794A1 (en) 2013-03-15 2016-05-20 Systems and methods for addressing a media database using distance associative hashing

Applications Claiming Priority (18)

Application Number Priority Date Filing Date Title
US201361791578P 2013-03-15 2013-03-15
US61/791,578 2013-03-15
US14/089,003 2013-11-25
US14/089,003 US8898714B2 (en) 2009-05-29 2013-11-25 Methods for identifying video segments and displaying contextually targeted content on a connected television
PCT/US2014/030795 WO2014145938A1 (en) 2013-03-15 2014-03-17 Systems and methods for real-time television ad detection using an automated content recognition database
US14/217,094 US8930980B2 (en) 2010-05-27 2014-03-17 Systems and methods for real-time television ad detection using an automated content recognition database
USPCT/US2014/30795 2014-03-17
US14/217,094 2014-03-17
USPCT/US2014/30805 2014-03-17
US14/217,375 2014-03-17
US14/217,375 US9094714B2 (en) 2009-05-29 2014-03-17 Systems and methods for on-screen graphics detection
US14/217,075 US9055309B2 (en) 2009-05-29 2014-03-17 Systems and methods for identifying video segments for displaying contextually relevant content
US14/217,435 US9094715B2 (en) 2009-05-29 2014-03-17 Systems and methods for multi-broadcast differentiation
PCT/US2014/030805 WO2014145947A1 (en) 2013-03-15 2014-03-17 Systems and methods for identifying video segments for displaying contextually relevant content
US14/217,425 US9071868B2 (en) 2009-05-29 2014-03-17 Systems and methods for improving server and client performance in fingerprint ACR systems
US14/217,425 2014-03-17
US14/217,435 2014-03-17
US14/217,075 2014-03-17

Publications (1)

Publication Number Publication Date
WO2014145929A1 true WO2014145929A1 (en) 2014-09-18

Family

ID=54258939

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2014/030782 WO2014145929A1 (en) 2013-03-15 2014-03-17 Systems and methods for addressing a media database using distance associative hashing

Country Status (7)

Country Link
CN (2) CN110083739B (en)
BR (1) BR112015023369B1 (en)
CA (1) CA2906199C (en)
CL (1) CL2015002621A1 (en)
HK (1) HK1217794A1 (en)
MX (2) MX356884B (en)
WO (1) WO2014145929A1 (en)

Cited By (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9154942B2 (en) 2008-11-26 2015-10-06 Free Stream Media Corp. Zero configuration communication between a browser and a networked media device
US9258383B2 (en) 2008-11-26 2016-02-09 Free Stream Media Corp. Monetization of television audience data across muliple screens of a user watching television
US9386356B2 (en) 2008-11-26 2016-07-05 Free Stream Media Corp. Targeting with television audience data across multiple screens
WO2016168556A1 (en) * 2015-04-17 2016-10-20 Vizio Inscape Technologies, Llc Systems and methods for reducing data density in large datasets
US9519772B2 (en) 2008-11-26 2016-12-13 Free Stream Media Corp. Relevancy improvement through targeting of information based on data gathered from a networked device associated with a security sandbox of a client device
WO2017011758A1 (en) * 2015-07-16 2017-01-19 Vizio Inscape Technologies, Llc Optimizing media fingerprint retention to improve system resource utilization
US9560425B2 (en) 2008-11-26 2017-01-31 Free Stream Media Corp. Remotely control devices over a network without authentication or registration
US9838753B2 (en) 2013-12-23 2017-12-05 Inscape Data, Inc. Monitoring individual viewing of television events using tracking pixels and cookies
US9906834B2 (en) 2009-05-29 2018-02-27 Inscape Data, Inc. Methods for identifying video segments and displaying contextually targeted content on a connected television
US9955192B2 (en) 2013-12-23 2018-04-24 Inscape Data, Inc. Monitoring individual viewing of television events using tracking pixels and cookies
US9961388B2 (en) 2008-11-26 2018-05-01 David Harrison Exposure of public internet protocol addresses in an advertising exchange server to improve relevancy of advertisements
US9986279B2 (en) 2008-11-26 2018-05-29 Free Stream Media Corp. Discovery, access control, and communication with networked services
US10116972B2 (en) 2009-05-29 2018-10-30 Inscape Data, Inc. Methods for identifying video segments and displaying option to view from an alternative source and/or on an alternative device
US10169455B2 (en) 2009-05-29 2019-01-01 Inscape Data, Inc. Systems and methods for addressing a media database using distance associative hashing
US10192138B2 (en) 2010-05-27 2019-01-29 Inscape Data, Inc. Systems and methods for reducing data density in large datasets
US10334324B2 (en) 2008-11-26 2019-06-25 Free Stream Media Corp. Relevant advertisement generation based on a user operating a client device communicatively coupled with a networked media device
US10375451B2 (en) 2009-05-29 2019-08-06 Inscape Data, Inc. Detection of common media segments
US10405014B2 (en) 2015-01-30 2019-09-03 Inscape Data, Inc. Methods for identifying video segments and displaying option to view from an alternative source and/or on an alternative device
US10419541B2 (en) 2008-11-26 2019-09-17 Free Stream Media Corp. Remotely control devices over a network without authentication or registration
US10567823B2 (en) 2008-11-26 2020-02-18 Free Stream Media Corp. Relevant advertisement generation based on a user operating a client device communicatively coupled with a networked media device
US10631068B2 (en) 2008-11-26 2020-04-21 Free Stream Media Corp. Content exposure attribution based on renderings of related content across multiple devices
US10873788B2 (en) 2015-07-16 2020-12-22 Inscape Data, Inc. Detection of common media segments
US10880340B2 (en) 2008-11-26 2020-12-29 Free Stream Media Corp. Relevancy improvement through targeting of information based on data gathered from a networked device associated with a security sandbox of a client device
US10902048B2 (en) 2015-07-16 2021-01-26 Inscape Data, Inc. Prediction of future views of video segments to optimize system resource utilization
US10949458B2 (en) 2009-05-29 2021-03-16 Inscape Data, Inc. System and method for improving work load management in ACR television monitoring system
US10977693B2 (en) 2008-11-26 2021-04-13 Free Stream Media Corp. Association of content identifier of audio-visual data with additional data through capture infrastructure
US10983984B2 (en) 2017-04-06 2021-04-20 Inscape Data, Inc. Systems and methods for improving accuracy of device maps using media viewing data
US11272248B2 (en) 2009-05-29 2022-03-08 Inscape Data, Inc. Methods for identifying video segments and displaying contextually targeted content on a connected television
US11308144B2 (en) 2015-07-16 2022-04-19 Inscape Data, Inc. Systems and methods for partitioning search indexes for improved efficiency in identifying media segments

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10970333B2 (en) * 2016-08-08 2021-04-06 NetraDyne, Inc. Distributed video storage and search with edge computing

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060155952A1 (en) * 2003-04-22 2006-07-13 Haas William R Memory management system and method using a hash table
US7346512B2 (en) * 2000-07-31 2008-03-18 Landmark Digital Services, Llc Methods for recognizing unknown media samples using characteristics of known media samples
US20080313140A1 (en) * 2007-06-18 2008-12-18 Zeitera, Llc Method and Apparatus for Multi-Dimensional Content Search and Video Identification
US20100306808A1 (en) * 2009-05-29 2010-12-02 Zeev Neumeier Methods for identifying video segments and displaying contextually targeted content on a connected television
US8094872B1 (en) * 2007-05-09 2012-01-10 Google Inc. Three-dimensional wavelet based video fingerprinting
US8171004B1 (en) * 2006-04-20 2012-05-01 Pinehill Technology, Llc Use of hash values for identification and location of content

Family Cites Families (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5812286A (en) * 1995-08-30 1998-09-22 Hewlett-Packard Company Automatic color processing to correct hue shift and incorrect exposure
US6064764A (en) * 1998-03-30 2000-05-16 Seiko Epson Corporation Fragile watermarks for detecting tampering in images
GB2351409B (en) * 1999-03-18 2003-10-22 British Broadcasting Corp Watermarking
US7089240B2 (en) * 2000-04-06 2006-08-08 International Business Machines Corporation Longest prefix match lookup using hash function
CN100431271C (en) * 2001-01-17 2008-11-05 皇家菲利浦电子有限公司 Robust checksums
US20030056010A1 (en) * 2001-09-20 2003-03-20 Koninklijke Philips Electronics N.V. Downstream metadata altering
WO2004074968A2 (en) * 2003-02-21 2004-09-02 Caringo, Inc. Additional hash functions in content-based addressing
US20050210501A1 (en) * 2004-03-19 2005-09-22 Microsoft Corporation Method and apparatus for handling metadata
US7469241B2 (en) * 2004-11-30 2008-12-23 Oracle International Corporation Efficient data aggregation operations using hash tables
US7685109B1 (en) * 2005-12-29 2010-03-23 Amazon Technologies, Inc. Method and apparatus for data partitioning and replication in a searchable data service
CN101473657A (en) * 2006-06-20 2009-07-01 皇家飞利浦电子股份有限公司 Generating fingerprints of video signals
WO2008143768A1 (en) * 2007-05-17 2008-11-27 Dolby Labortories Licensing Corporation Deriving video signatures that are insensitive to picture modification and frame-rate conversion
CN101162470B (en) * 2007-11-16 2011-04-20 北京交通大学 Video frequency advertisement recognition method based on layered matching
HUE026760T2 (en) * 2008-03-20 2016-07-28 Univ Geneve Secure item identification and authentication system and method based on unclonable features
GB2460844B (en) * 2008-06-10 2012-06-06 Half Minute Media Ltd Automatic detection of repeating video sequences
US8539199B2 (en) * 2010-03-12 2013-09-17 Lsi Corporation Hash processing in a network communications processor architecture
WO2010135082A1 (en) * 2009-05-19 2010-11-25 Dolby Laboratories Licensing Corporation Localized weak bit assignment
US8397028B2 (en) * 2010-06-15 2013-03-12 Stephen SPACKMAN Index entry eviction
EP2599295A1 (en) * 2010-07-30 2013-06-05 ByteMobile, Inc. Systems and methods for video cache indexing
CN102377960B (en) * 2010-08-24 2014-11-05 腾讯科技(深圳)有限公司 Video picture displaying method and device
US9110936B2 (en) * 2010-12-28 2015-08-18 Microsoft Technology Licensing, Llc Using index partitioning and reconciliation for data deduplication

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7346512B2 (en) * 2000-07-31 2008-03-18 Landmark Digital Services, Llc Methods for recognizing unknown media samples using characteristics of known media samples
US20060155952A1 (en) * 2003-04-22 2006-07-13 Haas William R Memory management system and method using a hash table
US8171004B1 (en) * 2006-04-20 2012-05-01 Pinehill Technology, Llc Use of hash values for identification and location of content
US8094872B1 (en) * 2007-05-09 2012-01-10 Google Inc. Three-dimensional wavelet based video fingerprinting
US20080313140A1 (en) * 2007-06-18 2008-12-18 Zeitera, Llc Method and Apparatus for Multi-Dimensional Content Search and Video Identification
US20100306808A1 (en) * 2009-05-29 2010-12-02 Zeev Neumeier Methods for identifying video segments and displaying contextually targeted content on a connected television

Cited By (65)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9838758B2 (en) 2008-11-26 2017-12-05 David Harrison Relevancy improvement through targeting of information based on data gathered from a networked device associated with a security sandbox of a client device
US9560425B2 (en) 2008-11-26 2017-01-31 Free Stream Media Corp. Remotely control devices over a network without authentication or registration
US9258383B2 (en) 2008-11-26 2016-02-09 Free Stream Media Corp. Monetization of television audience data across muliple screens of a user watching television
US9154942B2 (en) 2008-11-26 2015-10-06 Free Stream Media Corp. Zero configuration communication between a browser and a networked media device
US10986141B2 (en) 2008-11-26 2021-04-20 Free Stream Media Corp. Relevancy improvement through targeting of information based on data gathered from a networked device associated with a security sandbox of a client device
US9519772B2 (en) 2008-11-26 2016-12-13 Free Stream Media Corp. Relevancy improvement through targeting of information based on data gathered from a networked device associated with a security sandbox of a client device
US10977693B2 (en) 2008-11-26 2021-04-13 Free Stream Media Corp. Association of content identifier of audio-visual data with additional data through capture infrastructure
US9848250B2 (en) 2008-11-26 2017-12-19 Free Stream Media Corp. Relevancy improvement through targeting of information based on data gathered from a networked device associated with a security sandbox of a client device
US9576473B2 (en) 2008-11-26 2017-02-21 Free Stream Media Corp. Annotation of metadata through capture infrastructure
US9589456B2 (en) 2008-11-26 2017-03-07 Free Stream Media Corp. Exposure of public internet protocol addresses in an advertising exchange server to improve relevancy of advertisements
US9591381B2 (en) 2008-11-26 2017-03-07 Free Stream Media Corp. Automated discovery and launch of an application on a network enabled device
US9686596B2 (en) 2008-11-26 2017-06-20 Free Stream Media Corp. Advertisement targeting through embedded scripts in supply-side and demand-side platforms
US9706265B2 (en) 2008-11-26 2017-07-11 Free Stream Media Corp. Automatic communications between networked devices such as televisions and mobile devices
US9854330B2 (en) 2008-11-26 2017-12-26 David Harrison Relevancy improvement through targeting of information based on data gathered from a networked device associated with a security sandbox of a client device
US9716736B2 (en) 2008-11-26 2017-07-25 Free Stream Media Corp. System and method of discovery and launch associated with a networked media device
US10880340B2 (en) 2008-11-26 2020-12-29 Free Stream Media Corp. Relevancy improvement through targeting of information based on data gathered from a networked device associated with a security sandbox of a client device
US9386356B2 (en) 2008-11-26 2016-07-05 Free Stream Media Corp. Targeting with television audience data across multiple screens
US9167419B2 (en) 2008-11-26 2015-10-20 Free Stream Media Corp. Discovery and launch system and method
US9703947B2 (en) 2008-11-26 2017-07-11 Free Stream Media Corp. Relevancy improvement through targeting of information based on data gathered from a networked device associated with a security sandbox of a client device
US9866925B2 (en) 2008-11-26 2018-01-09 Free Stream Media Corp. Relevancy improvement through targeting of information based on data gathered from a networked device associated with a security sandbox of a client device
US10791152B2 (en) 2008-11-26 2020-09-29 Free Stream Media Corp. Automatic communications between networked devices such as televisions and mobile devices
US10771525B2 (en) 2008-11-26 2020-09-08 Free Stream Media Corp. System and method of discovery and launch associated with a networked media device
US9961388B2 (en) 2008-11-26 2018-05-01 David Harrison Exposure of public internet protocol addresses in an advertising exchange server to improve relevancy of advertisements
US9967295B2 (en) 2008-11-26 2018-05-08 David Harrison Automated discovery and launch of an application on a network enabled device
US9986279B2 (en) 2008-11-26 2018-05-29 Free Stream Media Corp. Discovery, access control, and communication with networked services
US10032191B2 (en) 2008-11-26 2018-07-24 Free Stream Media Corp. Advertisement targeting through embedded scripts in supply-side and demand-side platforms
US10074108B2 (en) 2008-11-26 2018-09-11 Free Stream Media Corp. Annotation of metadata through capture infrastructure
US10631068B2 (en) 2008-11-26 2020-04-21 Free Stream Media Corp. Content exposure attribution based on renderings of related content across multiple devices
US10567823B2 (en) 2008-11-26 2020-02-18 Free Stream Media Corp. Relevant advertisement generation based on a user operating a client device communicatively coupled with a networked media device
US10142377B2 (en) 2008-11-26 2018-11-27 Free Stream Media Corp. Relevancy improvement through targeting of information based on data gathered from a networked device associated with a security sandbox of a client device
US10425675B2 (en) 2008-11-26 2019-09-24 Free Stream Media Corp. Discovery, access control, and communication with networked services
US10419541B2 (en) 2008-11-26 2019-09-17 Free Stream Media Corp. Remotely control devices over a network without authentication or registration
US10334324B2 (en) 2008-11-26 2019-06-25 Free Stream Media Corp. Relevant advertisement generation based on a user operating a client device communicatively coupled with a networked media device
US10271098B2 (en) 2009-05-29 2019-04-23 Inscape Data, Inc. Methods for identifying video segments and displaying contextually targeted content on a connected television
US9906834B2 (en) 2009-05-29 2018-02-27 Inscape Data, Inc. Methods for identifying video segments and displaying contextually targeted content on a connected television
US10820048B2 (en) 2009-05-29 2020-10-27 Inscape Data, Inc. Methods for identifying video segments and displaying contextually targeted content on a connected television
US10949458B2 (en) 2009-05-29 2021-03-16 Inscape Data, Inc. System and method for improving work load management in ACR television monitoring system
US10375451B2 (en) 2009-05-29 2019-08-06 Inscape Data, Inc. Detection of common media segments
US11080331B2 (en) 2009-05-29 2021-08-03 Inscape Data, Inc. Systems and methods for addressing a media database using distance associative hashing
US10185768B2 (en) 2009-05-29 2019-01-22 Inscape Data, Inc. Systems and methods for addressing a media database using distance associative hashing
US10169455B2 (en) 2009-05-29 2019-01-01 Inscape Data, Inc. Systems and methods for addressing a media database using distance associative hashing
US11272248B2 (en) 2009-05-29 2022-03-08 Inscape Data, Inc. Methods for identifying video segments and displaying contextually targeted content on a connected television
US10116972B2 (en) 2009-05-29 2018-10-30 Inscape Data, Inc. Methods for identifying video segments and displaying option to view from an alternative source and/or on an alternative device
US10192138B2 (en) 2010-05-27 2019-01-29 Inscape Data, Inc. Systems and methods for reducing data density in large datasets
US9955192B2 (en) 2013-12-23 2018-04-24 Inscape Data, Inc. Monitoring individual viewing of television events using tracking pixels and cookies
US10306274B2 (en) 2013-12-23 2019-05-28 Inscape Data, Inc. Monitoring individual viewing of television events using tracking pixels and cookies
US11039178B2 (en) 2013-12-23 2021-06-15 Inscape Data, Inc. Monitoring individual viewing of television events using tracking pixels and cookies
US9838753B2 (en) 2013-12-23 2017-12-05 Inscape Data, Inc. Monitoring individual viewing of television events using tracking pixels and cookies
US10284884B2 (en) 2013-12-23 2019-05-07 Inscape Data, Inc. Monitoring individual viewing of television events using tracking pixels and cookies
US11711554B2 (en) 2015-01-30 2023-07-25 Inscape Data, Inc. Methods for identifying video segments and displaying option to view from an alternative source and/or on an alternative device
US10405014B2 (en) 2015-01-30 2019-09-03 Inscape Data, Inc. Methods for identifying video segments and displaying option to view from an alternative source and/or on an alternative device
US10945006B2 (en) 2015-01-30 2021-03-09 Inscape Data, Inc. Methods for identifying video segments and displaying option to view from an alternative source and/or on an alternative device
WO2016168556A1 (en) * 2015-04-17 2016-10-20 Vizio Inscape Technologies, Llc Systems and methods for reducing data density in large datasets
US10482349B2 (en) 2015-04-17 2019-11-19 Inscape Data, Inc. Systems and methods for reducing data density in large datasets
EP4375952A3 (en) * 2015-04-17 2024-06-19 Inscape Data, Inc. Systems and methods for reducing data density in large datasets
WO2017011758A1 (en) * 2015-07-16 2017-01-19 Vizio Inscape Technologies, Llc Optimizing media fingerprint retention to improve system resource utilization
US10902048B2 (en) 2015-07-16 2021-01-26 Inscape Data, Inc. Prediction of future views of video segments to optimize system resource utilization
US10873788B2 (en) 2015-07-16 2020-12-22 Inscape Data, Inc. Detection of common media segments
US10674223B2 (en) 2015-07-16 2020-06-02 Inscape Data, Inc. Optimizing media fingerprint retention to improve system resource utilization
US10080062B2 (en) 2015-07-16 2018-09-18 Inscape Data, Inc. Optimizing media fingerprint retention to improve system resource utilization
US11308144B2 (en) 2015-07-16 2022-04-19 Inscape Data, Inc. Systems and methods for partitioning search indexes for improved efficiency in identifying media segments
US11451877B2 (en) 2015-07-16 2022-09-20 Inscape Data, Inc. Optimizing media fingerprint retention to improve system resource utilization
US11659255B2 (en) 2015-07-16 2023-05-23 Inscape Data, Inc. Detection of common media segments
US11971919B2 (en) 2015-07-16 2024-04-30 Inscape Data, Inc. Systems and methods for partitioning search indexes for improved efficiency in identifying media segments
US10983984B2 (en) 2017-04-06 2021-04-20 Inscape Data, Inc. Systems and methods for improving accuracy of device maps using media viewing data

Also Published As

Publication number Publication date
CA2906199A1 (en) 2014-09-18
CA2906199C (en) 2021-08-24
MX2015012510A (en) 2015-12-16
HK1217794A1 (en) 2017-01-20
CN105144141B (en) 2018-12-07
BR112015023369B1 (en) 2022-04-05
CN110083739B (en) 2024-04-30
MX356884B (en) 2018-06-19
BR112015023369A2 (en) 2019-02-19
CL2015002621A1 (en) 2016-04-15
CN105144141A (en) 2015-12-09
CN110083739A (en) 2019-08-02
MX2020001441A (en) 2021-08-20

Similar Documents

Publication Publication Date Title
US11080331B2 (en) Systems and methods for addressing a media database using distance associative hashing
US9055335B2 (en) Systems and methods for addressing a media database using distance associative hashing
CA2906199C (en) Systems and methods for addressing a media database using distance associative hashing
EP3001871B1 (en) Systems and methods for addressing a media database using distance associative hashing
US20230289383A1 (en) Video fingerprinting
US9959345B2 (en) Search and identification of video content
KR102531622B1 (en) Prediction of future viewing of video segments to optimize system resource utilization
KR102711752B1 (en) System and method for dividing a search index for improved efficiency in identifying media segments
AU2016250276B2 (en) Systems and methods for reducing data density in large datasets
JP5980311B2 (en) Video signature
US9578394B2 (en) Video signature creation and matching
US20170091524A1 (en) Identifying video content via color-based fingerprint matching
US20160286266A1 (en) Labeling video content
US10015541B2 (en) Storing and retrieval heuristics
WO2016151415A1 (en) Storing and retrieval heuristics

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 201480017043.9

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14762850

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2906199

Country of ref document: CA

WWE Wipo information: entry into national phase

Ref document number: MX/A/2015/012510

Country of ref document: MX

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2014762850

Country of ref document: EP

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112015023369

Country of ref document: BR

REG Reference to national code

Ref country code: BR

Ref legal event code: B01E

Ref document number: 112015023369

Country of ref document: BR

Free format text: COMPROVE O DIREITO DE REIVINDICAR A PRIORIDADE US61/791,578 DE 15/03/2013 APRESENTANDO DOCUMENTO DE CESSAO CONTENDO OS DADOS DA PRIORIDADE E CEDIDO POR TODOS OS TITULARES, CONFORME A RESOLUCAO INPI/PR NO 179 DE 21/02/2017 NO ART 2O 1O, UMA VEZ QUE OS DOCUMENTOS DE CESSAO APRESENTADOS NA PETICAO 860150208388 NAO POSSUI O NUMERO DESSA PRIORIDADE E TEM COMO CEDENTE APENAS 2 DOS 4 TITULARES DA PRIORIDADE.

ENP Entry into the national phase

Ref document number: 112015023369

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20150914