CN104881642A - Method and device for content pushing, and equipment - Google Patents
Method and device for content pushing, and equipment Download PDFInfo
- Publication number
- CN104881642A CN104881642A CN201510267665.3A CN201510267665A CN104881642A CN 104881642 A CN104881642 A CN 104881642A CN 201510267665 A CN201510267665 A CN 201510267665A CN 104881642 A CN104881642 A CN 104881642A
- Authority
- CN
- China
- Prior art keywords
- human
- target body
- content
- target
- body object
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/174—Facial expression recognition
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
Embodiments of the invention disclose a method and a device for content pushing and equipment, relate to the technical field of display, and aim to improve accuracy of content recommendation. According to the embodiments of the invention, human body objects in a plurality of continuous video frames are detected, and dwell time of the human body objects during a preset time interval is determined according to positions of the human body objects in the plurality of continuous video frames; when the dwell time of the human body objects is longer than a preset time threshold, human body objects of which the dwell time is longer than the preset time threshold is determined as target human body objects, and human face information and human body contour information of the human body objects are detected; according to a preset model, the human face information and the human body contour information are matched respectively, and the attributes of the target human body objects are determined according to matching results; the proportion of the target human body objects of the same attribute in all the object human body objects is calculated, and the attribute of the largest proportion is determined; and contents corresponding to the attribute of the largest proportion are displayed. As a result, the abovementioned problem is solved.
Description
Technical field
The present invention relates to display technique field, particularly relate to a kind of content delivery method, device and equipment.
Background technology
Display device, also can be called display, display screen, screen etc., is the equipment of a kind of exportable image or thoughts and feelings information (such as Braille display that blind person designs).Growing Internet technology is that all kinds of display device provides abundanter, more intelligent function.
At present, display device can not only carry out alternately with network side, thus obtains the various information that pushed by network side, can also carry out alternately with user, thus provides information more targetedly for user.Be arranged at the public occasions such as the street, market, square due to this display device more, its towards customer group also there is very large uncertainty.Existing display device can be provided with face identification device, the user profile such as sex, age of the user utilizing face identification device to get, and the sex ratio in customer group and ratio of age are added up, according to statistics, get the propelling movement content of the customer group for large percentage, the customer group that may be large percentage due to this propelling movement content is most interested in, and therefore this mode can meet the demand of most of user preferably.
But the face information that existing this display device can identify is more single, only can identify the face of static state; Or when face is at least partially obscured, existing this display device just cannot get user profile accurately, thus have impact on the accuracy of commending contents.
Summary of the invention
The embodiment of the present invention provides a kind of content delivery method, device and equipment, in order to improve the precision of commending contents.
The embodiment of the present invention provides a kind of content delivery method, and the method comprises:
Detect continuously the human object in several frame of video, and according to the position of described human object in described several frame of video continuously, determine the described human object residence time within a preset time interval;
When the residence time of described human object is greater than preset time threshold, the human object being greater than described predetermined threshold value the residence time is defined as target body object, and detects face information and the human body contour outline information of described target body object;
According to preset model, described face information and human body contour outline information are mated respectively, according to matching result, determine the attribute of described target body object;
Statistics belongs to the ratio of target body object shared by target complete human object of same attribute, and determines the attribute that proportion is maximum;
Show the content that the maximum attribute of described proportion is corresponding.
The embodiment of the present invention also provides a kind of content pusher, and this device comprises:
Detecting unit, for detecting continuously the human object in several frame of video, and according to the position of described human object in described several frame of video continuously, determines the described human object residence time within a preset time interval;
Target determination unit, for when the residence time of described human object is greater than preset time threshold, the human object being greater than described predetermined threshold value the residence time is defined as target body object, and detects face information and the human body contour outline information of described target body object;
Analytic unit, for according to preset model, mates respectively to described face information and human body contour outline information, according to matching result, determines the attribute of described target body object;
Statistic unit, for adding up the target body object that belongs to same attribute ratio shared in target complete human object, and determines the attribute that proportion is maximum;
Push unit, for showing content corresponding to the maximum attribute of described proportion.
The embodiment of the present invention also provides a kind of display device, and this equipment comprises:
Imageing sensor, for according to default screening-mode, capture video frame;
Processor, for detecting continuously the human object in several frame of video, and according to the position of described human object in described several frame of video continuously, determines the described human object residence time within a preset time interval; When the residence time of described human object is greater than preset time threshold, the human object being greater than predetermined threshold value the residence time is defined as target body object, and detects face information and the human body contour outline information of described target body object; According to preset model, described face information and human body contour outline information are mated respectively, according to matching result, determine the attribute of described target body object; Statistics belongs to the ratio of target body object shared by target complete human object of same attribute, and determines the attribute that proportion is maximum; And will be used for representing that the picture signal of the content that the maximum attribute of described proportion is corresponding is sent to display screen;
Described display screen, for showing according to the picture signal sent by described processor.
As can be seen from technique scheme, first the embodiment of the present invention can detect continuously the human object in several frame of video, and according to the position of human object in several frame of video continuous, determines the human object residence time within a preset time interval; When the residence time of human object is greater than preset time threshold, the human object being greater than predetermined threshold value the residence time is defined as target body object; Thus define user to be analyzed; Then, detect face information and the human body contour outline information of target body object, the visible embodiment of the present invention can not only detect face information (as face width, highly, the essential characteristic such as face feature, skin color, hair zones), can also get human body contour outline information (as, the information such as height, shoulder breadth, measurements of the chest, waist and hips), identify according to the attribute such as sex, age of the above-mentioned information got to customer group and add up, thus improve the accuracy of identification, and then the accuracy pushing content for different user group can be improved.
Accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, below the accompanying drawing used required in describing embodiment is briefly introduced, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.
The side schematic view of a kind of display device that Fig. 1 provides for the embodiment of the present invention;
The schematic diagram of a kind of application scenarios that Fig. 2 provides for the embodiment of the present invention;
The schematic flow sheet of a kind of content delivery method that Fig. 3 provides for the embodiment of the present invention;
A kind of schematic flow sheet pushing content to designated user group that Fig. 4 provides for the embodiment of the present invention;
The structural representation of a kind of content pusher that Fig. 5 provides for the embodiment of the present invention;
The structural representation of a kind of display device that Fig. 6 provides for the embodiment of the present invention.
Embodiment
In order to make the object, technical solutions and advantages of the present invention clearly, below in conjunction with accompanying drawing, the present invention is described in further detail, and obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making other embodiments all obtained under creative work prerequisite, belong to the scope of protection of the invention.
The embodiment of the present invention can be applied in various display device, above-mentioned display device can include but not limited to equipment: intelligent television, computer, electronic bill-board, outdoor large LCDs, rear-projection etc., the embodiment of the present invention can also preferably be applicable to have in the display equipment of all kinds of external or built-in image collecting device, above-mentioned external or built-in image collecting device can include but not limited to as lower device: colour imagery shot, infrared pick-up head, 3D is (three-dimensional, three-dimensional) camera, the first-class shooting of depth camera or camera.
The embodiment of the present invention only by way of example to one preferably application scenarios be described.Fig. 1 shows the side schematic view of a kind of display device that the embodiment of the present invention provides, as shown in Figure 1, be provided with projection rear projection screen 11, at least one RGB (R is red, G is green, B is blue) colour imagery shot 12 and depth camera 13 can be provided with in the front of projection rear projection screen 11, the back side of projection rear projection screen 11 is provided with projector 14.Wherein, RGB colour imagery shot 12 and depth camera 13 may be used for the potential user group of catching in the crowd of flowing.Fig. 2 shows the schematic diagram of a kind of application scenarios that the embodiment of the present invention provides, as shown in Figure 2, be provided with imageing sensor 22 (i.e. image collecting device) on the top of display screen 21, this imageing sensor 22 is for catching the potential user group in the crowd of flowing.
Embodiments provide a kind of content delivery method towards various display device, Fig. 3 shows the schematic flow sheet of a kind of content delivery method that the embodiment of the present invention provides, and as shown in Figure 3, this flow process can comprise:
Step 31: detect continuously the human object in several frame of video, and according to the position of human object in several frame of video continuous, determine the human object residence time within a preset time interval.
Step 32: when the residence time of human object is greater than preset time threshold, the human object being greater than predetermined threshold value the residence time is defined as target body object, and detects face information and the human body contour outline information of target body object.
Step 33: according to preset model, mates respectively to face information and human body contour outline information, according to matching result, determines the attribute of target body object.
Step 34: statistics belongs to the ratio of target body object shared by target complete human object of same attribute, and determines the attribute that proportion is maximum.
Step 35: show the content that the maximum attribute of proportion is corresponding.
Optionally, after above-mentioned steps 35, detect the face characteristic change in each target body object and/or human body contour outline changing features; According to default characteristic model, the change of the face characteristic of each target body object and/or human body contour outline changing features are mated, according to matching result, determine each target body object whether being satisfied with the content of current display, and the ratio of target body object shared by target complete human object that statistics is satisfied, or, add up the ratio that unsatisfied target body object is shared in target complete human object; When the ratio that satisfied target body object is shared in target complete human object is more than or equal to the first predetermined threshold value, continue display Current Content; When the ratio that satisfied target body object is shared in target complete human object is less than the first predetermined threshold value, according to preset rules, change the content of current display; Or when the ratio that unsatisfied target body object is shared in target complete human object is greater than the second predetermined threshold value, according to preset rules, change the content of current display; When the ratio that unsatisfied target body object is shared in target complete human object is less than or equal to the second predetermined threshold value, continue display Current Content.
Optionally, in above-mentioned steps 33, obtain and preset face characteristic model; According to default face characteristic model, the face characteristic change of each target body object is mated, according to matching result, determines each target body object whether being satisfied with the content of current display.
Optionally, in above-mentioned steps 33, detect the change in location of each target body object in several frame of video respectively; According to the change in location detected, determine each target body object whether being satisfied with the content of current display.
Optionally, in above-mentioned steps 31, for each frame of video, detect the human body contour outline in this frame of video respectively, and reject the background interference in this frame of video; According to the situation of change of the human body contour outline in several frame of video continuous, determine the residence time of human object.
Below the embodiment of the present invention is described in detail.
The mode that the embodiment of the present invention adopts colour imagery shot to combine with depth camera, the common sex to personage, age are detected, not only can improve the accuracy of detection, but also can avoid because imageing sensor cannot detect face, and cause detecting failed problem.In addition, the embodiment of the present invention can also detect further and whether counting user (namely spectators) is satisfied with the content of current display, add and be directed to the concern of user to the satisfaction of content recommendation, thus can collect as follow-up propelling movement content provides the information of reference, and then content recommendation targetedly can be had more for user provides.
The embodiment of the present invention can by installing the detection of imageing sensor realization to the sex and age of mobile population additional to display device.The imageing sensor that the embodiment of the present invention adopts colour imagery shot to combine with depth camera, wherein, colour imagery shot can detect some essential characteristics such as the colours of skin of human body, and depth camera can realize the detection of height, measurements of the chest, waist and hips etc. to human body, the combination of two kinds of cameras can detect age and the sex of personage more accurately, and can avoid due to illumination or the phenomenon of detection failure that causes because can't detect face in conjunction with depth camera.
Fig. 4 shows a kind of schematic flow sheet pushing content to designated user group that the embodiment of the present invention provides, and as shown in Figure 4, this flow process can comprise:
Step 41: the human body contour outline in scene is detected, background is removed, and the human body contour outline detected is followed the tracks of, utilize the human body contour outline that the rectangle frame of different identification firmly identifies.
During specific implementation, with reference to Fig. 2, for the detection of different occasion customer group (spectators), first, the picture frame (i.e. frame of video) of whole scene is obtained by imageing sensor 22, the image that the image obtained for colour imagery shot and depth camera obtain all will carry out the segmentation of prospect and background, namely removes the interference of extraneous background.In embodiments of the present invention, first under suitable surround lighting, set up human body colouring information sample (comprising having: the positive sample of human body RGB information and negative sample) and human depth's message sample (comprising having: the positive sample of human depth's information and negative sample), what wherein the positive sample of human body RGB information was chosen is different by RGB camera collection posture, wear the picture of the different fixed size of human body under different background clothes, what wherein the negative sample of human body RGB information was chosen is without the picture of the fixed size of human body only containing different background, then the HOG feature (gradient orientation histogram of the HOG characteristic sum human body RGB negative sample of the positive sample of human body RGB is asked for respectively, Histogram of Oriented Gradient), finally utilize SVM (support vector machine, Support Vector Machine) Algorithm for Training draws HOG feature detection model.Positive sample and the negative sample of human depth's information gather by depth camera the picture that posture is different, wear the fixed size of different human bodies under different background clothes, then ask for the HOD feature of the HOD characteristic sum depth information negative sample of the positive sample of depth information respectively, finally utilize SVM Algorithm for Training to draw HOD feature detection model.
Because human body varies, need to use the detection that multiple dimensioned concept solves the pedestrian of different size in video window, simultaneously depth data figure heel distance between there is certain relation, therefore the large small scale that can be understood as the pedestrian within the scope of different distance is different, the embodiment of the present invention can accomplish depth data to be divided into several scopes according to the data of sample, each scope can the yardstick of a corresponding standard, the i.e. corresponding detection window of the personage of each different size, the RGB detection model of finally training out with SVM respectively and depth detection model judge that detection window is the probability of human body.The embodiment of the present invention also can identify an ID for each detection window being judged as human body.For example, continuing with reference to Fig. 2, after determining that the object in window is human object, is that each detection window identifies according to preset rules.
Step 42: detect the different colours residence time of rectangle frame in picture.
After obtaining the detection window of each personage, the embodiment of the present invention can adopt Camshift (MeanShift (cluster) algorithm of continuous adaptive, Continuously Adaptive Mean-SHIFT) track algorithm carries out tracking lock to person detecting window, and with different colours label detection window (with reference to Fig. 2, in embodiments of the present invention, can think window A, window B, window C and window D is all marked as different colors), ask for the barycenter of rectangle frame simultaneously, the coordinate of Real-time Obtaining rectangle barycenter in image, i.e. personage's motion, frame selects the rectangle frame of character contour also along with motion, the center-of-mass coordinate of rectangle frame changes in real time, by calculating coordinate and the figure image width of rectangle frame barycenter in real time, high relation, judge that the barycenter of rectangle frame is whether in image range, if the time that the barycenter of rectangle frame exists in image range is more than a time threshold T
set, just judge that the personage in rectangle frame rests on display device front, if the time that the barycenter of rectangle frame exists in image range is no more than T
set, just determine that the personage in rectangle frame does not stop in display device front.
Step 43: judge whether the residence time exceedes preset time threshold, if so, then performs step 44; Otherwise rebound performs step 41.
Step 44: plurality of human faces detection is carried out to the personage in rectangle frame.
During specific implementation, if judge that personage stops in display device front, depth camera will be used to carry out bone detection to human body, detect the trunk of human body and the head of human body, obtain the barycenter of human body head in depth image, and the three-dimensional coordinate of head barycenter is converted to two-dimensional coordinate, namely the position in the two dimensional image that obtains at RGB camera of human body head, then height h' and the width w' of human body head is obtained according to depth camera, in conjunction with the position of human body head barycenter in two dimensional image, the profile of human body head is drawn out in the two dimensional image that can obtain at RGB camera, the embodiment of the present invention can set a height threshold h according to the general thickness of personage's hair ", here using the minimum point in contouring head as initial point, h'-h " altitude range in be judged to be face scope, h " is judged to be the scope at human hair place in altitude range.Will calculate the color value of the pixel in this region at human hair scope internal program, namely the color of human hair is obtained by RGB camera, by carrying out color threshold segmentation to human body head, calculate the area of hair region, estimate the length of hair, the color of hair and length can provide certain foundation to the deduction of personage's sex and age.
Step 45: age bracket and the sex of inferring personage according to face characteristic.
During specific implementation, after obtaining the region of face, namely start to carry out Face datection.First the embodiment of the present invention can intercept the image in human face region and preserve, then compare with the face database template stored before, if the image in detection human face region does not possess the essential characteristic of face, namely judge that personage is back to screen, if the image in detection human face region possesses the essential characteristic of face, namely judge that personage is in the face of screen.If judge that personage is back to screen, just detected the color of personage's hair and the length of hair by colour imagery shot, simple sex and age of inferring personage, acquiescence is bob is the male sex, and long hair is women, and black is young, and white hair is old; If judge that personage is in the face of screen, not only to be detected the color of personage's hair and length by colour imagery shot, also to detect face characteristic, the sex and age of personage is inferred according to the feature of the width characteristics of face, altitude feature, eyes, nose, mouth, the colour of skin of its elderly and young people has certain difference, the skin of the elderly likely also have the features such as some spots, beard is also the important difference feature of age and sex, to the detection of personage's age and sex.
Step 46: in conjunction with the depth data that depth camera human body surface is different, determine age bracket and the sex of personage further.
During specific implementation, the embodiment of the present invention also combines depth camera and detects, according to the difference of men and women's measurements of the chest, waist and hips size, detect the measurements of the chest, waist and hips of personage, specific practice draws out human body contour outline in depth image, ask for the depth value that in human body contour outline, each pixel is corresponding, according to the position of head and trunk, substantially delimit the region at chest place, and calculate the depth value of each pixel in chest region, then the thickness of human body by obtaining, estimates the information such as the chest measurement of human body, shoulder breadth, hip circumference; According to the difference of adult and child height, the embodiment of the present invention detects the height of personage by depth camera, the information that the information obtained by colour imagery shot and depth camera obtain combines, the relatively accurate sex and age of inferring personage, and the sex and age section that simple dependence detection face can be avoided to cause detects failed phenomenon.
Like this, the embodiment of the present invention can not only utilize human face detection tech, the sex, age etc. of target body is identified by face characteristic, further, can also utilize and strengthen vision (computer vision) technology, obtain the information of target body more accurately, for example: utilize the three-dimensional information of target body and in conjunction with face characteristic information determination sex, utilize the height information of target body and determine the age etc. in conjunction with face characteristic information; Visible, the embodiment of the present invention can identify the attribute of target body object according to more various information, makes the result that identifies more accurate, thus can push content more accurately for user provides.
Step 47: judge whether the age of the personage in each rectangle frame and sex are determined, if so, then performs step 48; Otherwise rebound performs step 44.
Step 48: the ratio of age in statistics crowd and sex ratio.
During specific implementation, after image detection and process complete, the embodiment of the present invention can be added up to the sex and age ratio of personage in crowd.For example, in embodiments of the present invention, age bracket is divided into old age, youth, children substantially.
Step 49: push the larger interested displaying contents of crowd of accounting according to ratio of age and sex ratio proportion.
Step 410: after the content that display pushes, continues to detect personage's eye, nose, mouth feature change.
Step 411: according to personage's eye, nose, mouth feature change, personage is to the satisfaction of displaying contents for statistics.
Step 412: judge whether the satisfaction in crowd is more than or equal to 50%, if so, then performs step 413; Otherwise, perform step 414.
It should be noted that, whether above-mentioned satisfaction is more than or equal to 50% only as an example, in actual applications, can arrange this threshold value according to actual conditions.
Step 413: continue to push current displaying contents.
Step 414: according to preset rules, after changing the content of current display, rebound performs step 410.
For example, if elderly men accounting in crowd is higher, control module will be sent a signal to, push the interested displaying contents of elderly men, imageing sensor now in system will detect again to everyone face in crowd, according to the eye of face, nose, the change of mouth feature, infer the expression shape change of personage, namely judge whether spectators are satisfied with the content of current broadcasting or interested, by detecting that human face expression change obtains beholder indirectly to level of interest or the satisfaction of throwing in content, the specific aim and accuracy of throwing in content can be improved, also improve the intelligence degree of throwing in content device.Below by concrete example, specific practice is the seven kinds of expression picture storehouses preparing personage, these seven kinds expressions are anger (anger) respectively, disgust (dislike), fear (fear), happy (happily), sadness (sadness), surprise (pleasantly surprised), a large amount of pictures of different personage this kind of expression are all stored in often kind of expression storehouse, the face characteristic of the spectators caught and the human face expression feature in expression storehouse are carried out similarity mode by system, if it is high to detect expression picture matching degree in the expressive features of spectators and anger, so illustrate that spectators are unsatisfied with the content of current broadcasting or lose interest in, the expression of each personage express one's feelings with seven kinds successively in this way in picture carry out similarity mode, judge whether spectators are satisfied with current play content.
If the face of spectators is blocked, system also can detect personage's change in location in the picture by depth camera, infer according to convention, if spectators are unsatisfied with the content of current broadcasting or lose interest in, the time that so spectators stop in display screen front will significantly shorten, even can turn round at once and leave, so after displaying contents is play, if colour imagery shot does not capture face, so depth camera can detect the change of the spacing of personage and display screen in the very first time, detect the change of personage position in the picture simultaneously, if the distance detecting the distance of personage and display screen is more and more far away, so just judge that the detected content of personage to current broadcasting is lost interest in or be unsatisfied with.When each person detecting in system on human group is complete, system will be unsatisfied with the quantity of spectators to current play content in statistics crowd, then the accounting of satisfaction in calculating crowd, if ratio satisfied in crowd is more than or equal to a proportion threshold value P%, so system plays current content by continuing, if ratio satisfied in crowd is less than proportion threshold value P%, so again higher according to other age-sex's accountings except elderly men colony is carried out content propelling movement by system, then continue to detect the satisfaction of crowd, until the satisfaction ratio of crowd is more than or equal to proportion threshold value P%.
Again for example, if young man accounting in crowd is higher, control module will be sent a signal to, push the interested displaying contents of young man, imageing sensor now in system will detect again to everyone face in crowd, according to the eye of face, nose, the change of mouth feature, infer the expression shape change of personage, namely judge whether spectators are satisfied with the content of current broadcasting or interested, specific practice is the seven kinds of expression picture storehouses preparing personage, these seven kinds expressions are anger (anger) respectively, disgust (dislike), fear (fear), happy (happily), sadness (sadness), surprise (pleasantly surprised), a large amount of pictures of different personage this kind of expression are all stored in often kind of expression storehouse, the face characteristic of the spectators caught and the human face expression feature in expression storehouse are carried out similarity mode by system, if it is high to detect expression picture matching degree in the expressive features of spectators and anger, so illustrate that spectators are unsatisfied with the content of current broadcasting or lose interest in, the expression of each personage express one's feelings with seven kinds successively in this way in picture carry out similarity mode, judge whether spectators are satisfied with current play content.If the face of spectators is blocked, system also can detect personage's change in location in the picture by depth camera, infer according to convention, if spectators are unsatisfied with the content of current broadcasting or lose interest in, the time that so spectators stop in display screen front will significantly shorten, even can turn round at once and leave, so after displaying contents is play, if colour imagery shot does not capture face, so depth camera can detect the change of the spacing of personage and display screen in the very first time, detect the change of personage position in the picture simultaneously, if the distance detecting the distance of personage and display screen is more and more far away, so just judge that the detected content of personage to current broadcasting is lost interest in or be unsatisfied with.When each person detecting in system on human group is complete, system will be unsatisfied with the quantity of spectators to current play content in statistics crowd, then the accounting of satisfaction in calculating crowd, if ratio satisfied in crowd is more than or equal to proportion threshold value P%, so system plays current content by continuing, if ratio satisfied in crowd is less than proportion threshold value P%, so again higher according to other age-sex's accountings except young man colony is carried out content propelling movement by system, then continue to detect the satisfaction of crowd, until the satisfaction ratio of crowd is more than or equal to proportion threshold value P%.
Again for example, if boy's accounting in crowd is higher, control module will be sent a signal to, push the interested displaying contents of boy, imageing sensor now in system will detect again to everyone face in crowd, according to the eye of face, nose, the change of mouth feature, infer the expression shape change of personage, namely judge whether spectators are satisfied with the content of current broadcasting or interested, specific practice is the seven kinds of expression picture storehouses preparing personage, these seven kinds expressions are anger (anger) respectively, disgust (dislike), fear (fear), happy (happily), sadness (sadness), surprise (pleasantly surprised), a large amount of pictures of different personage this kind of expression are all stored in often kind of expression storehouse, the face characteristic of the spectators caught and the human face expression feature in expression storehouse are carried out similarity mode by system, if it is high to detect expression picture matching degree in the expressive features of spectators and anger, so illustrate that spectators are unsatisfied with the content of current broadcasting or lose interest in, the expression of each personage express one's feelings with seven kinds successively in this way in picture carry out similarity mode, judge whether spectators are satisfied with current play content.If the face of spectators is blocked, system also can detect personage's change in location in the picture by depth camera, infer according to convention, if spectators are unsatisfied with the content of current broadcasting or lose interest in, the time that so spectators stop in display screen front will significantly shorten, even can turn round at once and leave, so after displaying contents is play, if colour imagery shot does not capture face, so depth camera can detect the change of the spacing of personage and display screen in the very first time, detect the change of personage position in the picture simultaneously, if the distance detecting the distance of personage and display screen is more and more far away, so just judge that the detected content of personage to current broadcasting is lost interest in or be unsatisfied with.When each person detecting in system on human group is complete, system will be unsatisfied with the quantity of spectators to current play content in statistics crowd, then the accounting of satisfaction in calculating crowd, if ratio satisfied in crowd is more than or equal to proportion threshold value P%, so system plays current content by continuing, if ratio satisfied in crowd is less than proportion threshold value P%, so again higher according to other age-sex's accountings except boy colony is carried out content propelling movement by system, then continue to detect the satisfaction of crowd, until the satisfaction ratio of crowd is more than or equal to proportion threshold value P%.
Again for example, if old women accounting in crowd is higher, control module will be sent a signal to, push the interested displaying contents of old women, imageing sensor now in system will detect again to everyone face in crowd, according to the eye of face, nose, the change of mouth feature, infer the expression shape change of personage, namely judge whether spectators are satisfied with the content of current broadcasting or interested, specific practice is the seven kinds of expression picture storehouses preparing personage, these seven kinds expressions are anger (anger) respectively, disgust (dislike), fear (fear), happy (happily), sadness (sadness), surprise (pleasantly surprised), a large amount of pictures of different personage this kind of expression are all stored in often kind of expression storehouse, the face characteristic of the spectators caught and the human face expression feature in expression storehouse are carried out similarity mode by system, if it is high to detect expression picture matching degree in the expressive features of spectators and anger, so illustrate that spectators are unsatisfied with the content of current broadcasting or lose interest in, the expression of each personage express one's feelings with seven kinds successively in this way in picture carry out similarity mode, judge whether spectators are satisfied with current play content.If the face of spectators is blocked, system also can detect personage's change in location in the picture by depth camera, infer according to convention, if spectators are unsatisfied with the content of current broadcasting or lose interest in, the time that so spectators stop in display screen front will significantly shorten, even can turn round at once and leave, so after displaying contents is play, if colour imagery shot does not capture face, so depth camera can detect the change of the spacing of personage and display screen in the very first time, detect the change of personage position in the picture simultaneously, if the distance detecting the distance of personage and display screen is more and more far away, so just judge that the detected content of personage to current broadcasting is lost interest in or be unsatisfied with.When each person detecting in system on human group is complete, system will be unsatisfied with the quantity of spectators to current play content in statistics crowd, then the accounting of satisfaction in calculating crowd, if ratio satisfied in crowd is more than or equal to proportion threshold value P%, so system plays current content by continuing, if ratio satisfied in crowd is less than proportion threshold value P%, so again higher according to other age-sex's accountings except old women colony is carried out content propelling movement by system, then continue to detect the satisfaction of crowd, until the satisfaction ratio of crowd is more than or equal to proportion threshold value P%.
Again for example, if young women accounting in crowd is higher, control module will be sent a signal to, push the interested displaying contents of young women, imageing sensor now in system will detect again to everyone face in crowd, according to the eye of face, nose, the change of mouth feature, infer the expression shape change of personage, namely judge whether spectators are satisfied with the content of current broadcasting or interested, specific practice is the seven kinds of expression picture storehouses preparing personage, these seven kinds expressions are anger (anger) respectively, disgust (dislike), fear (fear), happy (happily), sadness (sadness), surprise (pleasantly surprised), a large amount of pictures of different personage this kind of expression are all stored in often kind of expression storehouse, the face characteristic of the spectators caught and the human face expression feature in expression storehouse are carried out similarity mode by system, if it is high to detect expression picture matching degree in the expressive features of spectators and anger, so illustrate that spectators are unsatisfied with the content of current broadcasting or lose interest in, the expression of each personage express one's feelings with seven kinds successively in this way in picture carry out similarity mode, judge whether spectators are satisfied with current play content.If the face of spectators is blocked, system also can detect personage's change in location in the picture by depth camera, infer according to convention, if spectators are unsatisfied with the content of current broadcasting or lose interest in, the time that so spectators stop in display screen front will significantly shorten, even can turn round at once and leave, so after displaying contents is play, if colour imagery shot does not capture face, so depth camera can detect the change of the spacing of personage and display screen in the very first time, detect the change of personage position in the picture simultaneously, if the distance detecting the distance of personage and display screen is more and more far away, so just judge that the detected content of personage to current broadcasting is lost interest in or be unsatisfied with.When each person detecting in system on human group is complete, system will be unsatisfied with the quantity of spectators to current play content in statistics crowd, then the accounting of satisfaction in calculating crowd, if ratio satisfied in crowd is more than or equal to proportion threshold value P%.So system plays current content by continuing, if ratio satisfied in crowd is less than proportion threshold value P%, so again higher according to other age-sex's accountings except young women colony is carried out content propelling movement by system, then continue to detect the satisfaction of crowd, until the satisfaction ratio of crowd is more than or equal to proportion threshold value P%.
Again for example, if girl's accounting in crowd is higher, control module will be sent a signal to, push the interested displaying contents of girl, imageing sensor now in system will detect again to everyone face in crowd, according to the eye of face, nose, the change of mouth feature, infer the expression shape change of personage, namely judge whether spectators are satisfied with the content of current broadcasting or interested, specific practice is the seven kinds of expression picture storehouses preparing personage, these seven kinds expressions are anger (anger) respectively, disgust (dislike), fear (fear), happy (happily), sadness (sadness), surprise (pleasantly surprised), a large amount of pictures of different personage this kind of expression are all stored in often kind of expression storehouse, the face characteristic of the spectators caught and the human face expression feature in expression storehouse are carried out similarity mode by system, if it is high to detect expression picture matching degree in the expressive features of spectators and anger, so illustrate that spectators are unsatisfied with the content of current broadcasting or lose interest in, the expression of each personage express one's feelings with seven kinds successively in this way in picture carry out similarity mode, judge whether spectators are satisfied with current play content.If the face of spectators is blocked, system also can detect personage's change in location in the picture by depth camera, infer according to convention, if spectators are unsatisfied with the content of current broadcasting or lose interest in, the time that so spectators stop in display screen front will significantly shorten, even can turn round at once and leave, so after displaying contents is play, if colour imagery shot does not capture face, so depth camera can detect the change of the spacing of personage and display screen in the very first time, detect the change of personage position in the picture simultaneously, if the distance detecting the distance of personage and display screen is more and more far away, so just judge that the detected content of personage to current broadcasting is lost interest in or be unsatisfied with.When each person detecting in system on human group is complete, system will be unsatisfied with the quantity of spectators to current play content in statistics crowd, then the accounting of satisfaction in calculating crowd, if ratio satisfied in crowd is more than or equal to proportion threshold value P%, so system plays current content by continuing, if ratio satisfied in crowd is less than proportion threshold value P%, so again higher according to other age-sex's accountings except girl colony is carried out content propelling movement by system, then continue to detect the satisfaction of crowd, until the satisfaction ratio of crowd is more than or equal to proportion threshold value P%.
Visible, first the embodiment of the present invention can detect continuously the human object in several frame of video, and according to the position of human object in several frame of video continuous, determines the human object residence time within a preset time interval; When the residence time of human object is greater than preset time threshold, the human object being greater than predetermined threshold value the residence time is defined as target body object; Thus define user to be analyzed; Then, detect face information and the human body contour outline information of target body object, the visible embodiment of the present invention can not only detect face information (as face width, highly, the essential characteristic such as face feature, skin color, hair zones), can also get human body contour outline information (as, the information such as height, shoulder breadth, measurements of the chest, waist and hips), identify according to the attribute such as sex, age of the above-mentioned information got to customer group and add up, thus improve the accuracy of identification, and then the accuracy pushing content for different user group can be improved.Further, the embodiment of the present invention is on the basis of precision that ensure that content recommendation, also add a kind of scheme detecting user satisfaction, can by detecting the mode of face characteristic change and/or human body contour outline changing features, judge the satisfaction of each target body to the content of current display, and then the content that needs to change current display is judged whether according to the satisfaction of most of user, change the traditional approach of the classification only being pushed content by network side unilateral decision, like this, embodiments provide a kind of two-way push mode, can not only be pushed to user according to up-to-date content by network side server, the feedback of user to present displayed content can also be obtained, and then achieve the higher push mode of a kind of interactive degree, propelling movement content is made to have more specific aim, also propelling movement mode can be made more intelligent, thus improve Consumer's Experience.
It should be noted that, the embodiment of the present invention is only illustrated with above-mentioned embodiment, the method that the embodiment of the present invention provides, device and equipment can according to all kinds of characteristics of human body (being not limited to the essential characteristics such as the colour of skin, height, hair color, hair zones, measurements of the chest, waist and hips information), the embodiment of the present invention even can also coordinate other computer vision means such as iris diagnostic device, 3D filming apparatus, further accuracy and the specific aim improving propelling movement content, repeats no more here.
Also it should be noted that, under the embodiment of the present invention is particularly useful for public arena, such as Projection Display advertisement screen, be placed in the large-scale rear-projection equipment and LCD screen etc. on square, illustrate no longer one by one here.
Based on identical technical conceive, the embodiment of the present invention also provides a kind of content pusher, and Fig. 5 shows the structural representation of a kind of content pusher that the embodiment of the present invention provides, and as shown in Figure 5, this device comprises:
Detecting unit 51, for detecting continuously the human object in several frame of video, and according to the position of described human object in described several frame of video continuously, determines the described human object residence time within a preset time interval;
Target determination unit 52, for when the residence time of described human object is greater than preset time threshold, the human object being greater than predetermined threshold value the residence time is defined as target body object, and detects face information and the human body contour outline information of described target body object;
Analytic unit 53, for according to preset model, mates respectively to described face information and human body contour outline information, according to matching result, determines the attribute of described target body object;
Statistic unit 54, for adding up the target body object that belongs to same attribute ratio shared in target complete human object, and determines the attribute that proportion is maximum;
Push unit 55, for showing content corresponding to the maximum attribute of described proportion.
Optionally, described target determination unit 52 also for:
After described push unit 55 shows content corresponding to the maximum attribute of described proportion, detect the face characteristic change in each target body object and/or human body contour outline changing features;
Described analytic unit 53 also for: according to default characteristic model, the change of the face characteristic of each target body object described and/or human body contour outline changing features are mated, according to matching result, determine each target body object whether being satisfied with the content of current display, and the ratio of target body object shared by target complete human object that statistics is satisfied, or, add up the ratio that unsatisfied target body object is shared in target complete human object;
Described push unit 55 also for: when the ratio that the target body object of described satisfaction is shared in target complete human object is more than or equal to the first predetermined threshold value, continue display Current Content; When the ratio that the target body object of described satisfaction is shared in target complete human object is less than described first predetermined threshold value, according to preset rules, change the content of current display; Or when the ratio that described unsatisfied target body object is shared in target complete human object is greater than the second predetermined threshold value, according to preset rules, change the content of current display; When the ratio that described unsatisfied target body object is shared in target complete human object is less than or equal to described second predetermined threshold value, continue display Current Content.
Optionally, described analytic unit 53 specifically for: obtain preset face characteristic model; According to described default face characteristic model, the face characteristic change of each target body object described is mated, according to matching result, determines each target body object whether being satisfied with the content of current display.
Optionally, described analytic unit 53 specifically for: detect the described change in location of each target body object in several frame of video respectively; According to the change in location detected, determine each target body object whether being satisfied with the content of current display.
Optionally, described detecting unit 51 specifically for: for each frame of video, detect the human body contour outline in this frame of video respectively, and reject the background interference in this frame of video; According to the situation of change of the human body contour outline in several frame of video continuous, determine the residence time of described human object.
Based on identical technical conceive, the embodiment of the present invention also provides a kind of display device, and Fig. 6 shows the structural representation of a kind of display device that the embodiment of the present invention provides, and as shown in Figure 6, this equipment comprises:
Imageing sensor 61, for according to default screening-mode, capture video frame;
Processor 62, for detecting continuously the human object in several frame of video, and according to the position of described human object in described several frame of video continuously, determines the described human object residence time within a preset time interval; When the residence time of described human object is greater than preset time threshold, the human object being greater than predetermined threshold value the residence time is defined as target body object, and detects face information and the human body contour outline information of described target body object; According to preset model, described face information and human body contour outline information are mated respectively, according to matching result, determine the attribute of described target body object; Statistics belongs to the ratio of target body object shared by target complete human object of same attribute, and determines the attribute that proportion is maximum; And will be used for representing that the picture signal of the content that the maximum attribute of described proportion is corresponding is sent to display screen 63;
Described display screen 63, for showing according to the picture signal sent by described processor 62.
Optionally, described processor 62 also for: after described display screen 63 shows content corresponding to the maximum attribute of described proportion, detect in each target body object face characteristic change and/or human body contour outline changing features; According to default characteristic model, the change of the face characteristic of each target body object described and/or human body contour outline changing features are mated, according to matching result, determine each target body object whether being satisfied with the content of current display, and the ratio of target body object shared by target complete human object that statistics is satisfied, or, add up the ratio that unsatisfied target body object is shared in target complete human object; When the ratio that the target body object of described satisfaction is shared in target complete human object is more than or equal to the first predetermined threshold value, continue display Current Content; When the ratio that the target body object of described satisfaction is shared in target complete human object is less than described first predetermined threshold value, according to preset rules, change the content of current display; Or when the ratio that described unsatisfied target body object is shared in target complete human object is greater than the second predetermined threshold value, according to preset rules, change the content of current display; When the ratio that described unsatisfied target body object is shared in target complete human object is less than or equal to described second predetermined threshold value, continue display Current Content.
Optionally, described processor 62 specifically for: obtain preset face characteristic model;
According to described default face characteristic model, the face characteristic change of each target body object described is mated, according to matching result, determines each target body object whether being satisfied with the content of current display.
Optionally, described processor 62 specifically for: detect the described change in location of each target body object in several frame of video respectively; According to the change in location detected, determine each target body object whether being satisfied with the content of current display.
Optionally, described processor 62 specifically for: for each frame of video, detect the human body contour outline in this frame of video respectively, and reject the background interference in this frame of video;
According to the change in location situation of the human body contour outline in several frame of video continuous, determine the described human object residence time within a preset time interval.
The present invention describes with reference to according to the process flow diagram of the method for the embodiment of the present invention, equipment (system) and computer program and/or block scheme.Should understand can by the combination of the flow process in each flow process in computer program instructions realization flow figure and/or block scheme and/or square frame and process flow diagram and/or block scheme and/or square frame.These computer program instructions can be provided to the processor of multi-purpose computer, special purpose computer, Embedded Processor or other programmable data processing device, make the function that the instruction that performed by the processor of this computing machine or other programmable data processing device can be specified in a flow process in realization flow figure or multiple flow process and/or block scheme square frame or multiple square frame.
These computer program instructions also can be stored in can in the computer-readable memory that works in a specific way of vectoring computer or other programmable data processing device, the instruction making to be stored in this computer-readable memory produces the manufacture comprising command device, and this command device realizes the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.
These computer program instructions also can be loaded in computing machine or other programmable data processing device, make on computing machine or other programmable devices, to perform sequence of operations step to produce computer implemented process, thus the instruction performed on computing machine or other programmable devices is provided for the step realizing the function of specifying in a flow process of process flow diagram or a square frame of multiple flow process and/or block scheme or multiple square frame.
Although describe the preferred embodiments of the present invention, those skilled in the art once obtain the basic creative concept of cicada, then can make other change and amendment to these embodiments.So claims are intended to be interpreted as comprising preferred embodiment and falling into all changes and the amendment of the scope of the invention.
Obviously, those skilled in the art can carry out various change and modification to the present invention and not depart from the spirit and scope of the present invention.Like this, if these amendments of the present invention and modification belong within the scope of the claims in the present invention and equivalent technologies thereof, then the present invention is also intended to comprise these change and modification.
Claims (11)
1. a content delivery method, is characterized in that, the method comprises:
Detect continuously the human object in several frame of video, and according to the position of described human object in described several frame of video continuously, determine the described human object residence time within a preset time interval;
When the residence time of described human object is greater than preset time threshold, the human object being greater than described predetermined threshold value the residence time is defined as target body object, and detects face information and the human body contour outline information of described target body object;
According to preset model, described face information and human body contour outline information are mated respectively, according to matching result, determine the attribute of described target body object;
Statistics belongs to the ratio of target body object shared by target complete human object of same attribute, and determines the attribute that proportion is maximum;
Show the content that the maximum attribute of described proportion is corresponding.
2. the method for claim 1, is characterized in that, after the content that the attribute that the described proportion of display is maximum is corresponding, the method also comprises:
Detect the face characteristic change in each target body object and/or human body contour outline changing features;
According to default characteristic model, the change of the face characteristic of each target body object described and/or human body contour outline changing features are mated, according to matching result, determine each target body object whether being satisfied with the content of current display, and the ratio of target body object shared by target complete human object that statistics is satisfied, or, add up the ratio that unsatisfied target body object is shared in target complete human object;
When the ratio that the target body object of described satisfaction is shared in target complete human object is more than or equal to the first predetermined threshold value, continue display Current Content; When the ratio that the target body object of described satisfaction is shared in target complete human object is less than described first predetermined threshold value, according to preset rules, change the content of current display; Or
When the ratio that described unsatisfied target body object is shared in target complete human object is greater than the second predetermined threshold value, according to preset rules, change the content of current display; When the ratio that described unsatisfied target body object is shared in target complete human object is less than or equal to described second predetermined threshold value, continue display Current Content.
3. method as claimed in claim 2, it is characterized in that, described basis presets characteristic model, the face characteristic change of each target body object described is mated, according to matching result, determine each target body object whether being satisfied with the content of current display, be specially:
Obtain and preset face characteristic model;
According to described default face characteristic model, the face characteristic change of each target body object described is mated, according to matching result, determines each target body object whether being satisfied with the content of current display.
4. method as claimed in claim 2, it is characterized in that, the described human body contour outline changing features to each target body object described is mated, and according to matching result, determines each target body object whether being satisfied with the content of current display, is specially:
Detect the described change in location of each target body object in several frame of video respectively; According to the change in location detected, determine each target body object whether being satisfied with the content of current display.
5. the method according to any one of claim 1-4, it is characterized in that, human object in described detection several frame of video continuous, and according to the position of described human object in described several frame of video continuously, determine the described human object residence time within a preset time interval; Be specially:
For each frame of video, detect the human body contour outline in this frame of video respectively, and reject the background interference in this frame of video;
According to the change in location situation of the human body contour outline in several frame of video continuous, determine the described human object residence time within a preset time interval.
6. a content pusher, is characterized in that, this device comprises:
Detecting unit, for detecting continuously the human object in several frame of video, and according to the position of described human object in described several frame of video continuously, determines the described human object residence time within a preset time interval;
Target determination unit, for when the residence time of described human object is greater than preset time threshold, the human object being greater than described predetermined threshold value the residence time is defined as target body object, and detects face information and the human body contour outline information of described target body object;
Analytic unit, for according to preset model, mates respectively to described face information and human body contour outline information, according to matching result, determines the attribute of described target body object;
Statistic unit, for adding up the target body object that belongs to same attribute ratio shared in target complete human object, and determines the attribute that proportion is maximum;
Push unit, for showing content corresponding to the maximum attribute of described proportion.
7. device as claimed in claim 6, is characterized in that, described target determination unit also for:
After described push unit shows content corresponding to the maximum attribute of described proportion, detect the face characteristic change in each target body object and/or human body contour outline changing features;
Described analytic unit also for: according to default characteristic model, the change of the face characteristic of each target body object described and/or human body contour outline changing features are mated, according to matching result, determine each target body object whether being satisfied with the content of current display, and the ratio of target body object shared by target complete human object that statistics is satisfied, or, add up the ratio that unsatisfied target body object is shared in target complete human object;
Described push unit also for: when the ratio that the target body object of described satisfaction is shared in target complete human object is more than or equal to the first predetermined threshold value, continue display Current Content; When the ratio that the target body object of described satisfaction is shared in target complete human object is less than described first predetermined threshold value, according to preset rules, change the content of current display; Or when the ratio that described unsatisfied target body object is shared in target complete human object is greater than the second predetermined threshold value, according to preset rules, change the content of current display; When the ratio that described unsatisfied target body object is shared in target complete human object is less than or equal to described second predetermined threshold value, continue display Current Content.
8. device as claimed in claim 7, is characterized in that, described analytic unit specifically for:
Obtain and preset face characteristic model;
According to described default face characteristic model, the face characteristic change of each target body object described is mated, according to matching result, determines each target body object whether being satisfied with the content of current display.
9. device as claimed in claim 7, is characterized in that, described analytic unit specifically for:
Detect the described change in location of each target body object in several frame of video respectively; According to the change in location detected, determine each target body object whether being satisfied with the content of current display.
10. the device according to any one of claim 6-9, is characterized in that, described detecting unit specifically for:
For each frame of video, detect the human body contour outline in this frame of video respectively, and reject the background interference in this frame of video;
According to the change in location situation of the human body contour outline in several frame of video continuous, determine the described human object residence time within a preset time interval.
11. 1 kinds of display devices, is characterized in that, this equipment comprises:
Imageing sensor, for according to default screening-mode, capture video frame;
Processor, for detecting continuously the human object in several frame of video, and according to the position of described human object in described several frame of video continuously, determines the described human object residence time within a preset time interval; When the residence time of described human object is greater than preset time threshold, the human object being greater than predetermined threshold value the residence time is defined as target body object, and detects face information and the human body contour outline information of described target body object; According to preset model, described face information and human body contour outline information are mated respectively, according to matching result, determine the attribute of described target body object; Statistics belongs to the ratio of target body object shared by target complete human object of same attribute, and determines the attribute that proportion is maximum; And will be used for representing that the picture signal of the content that the maximum attribute of described proportion is corresponding is sent to display screen;
Described display screen, for showing according to the picture signal sent by described processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510267665.3A CN104881642B (en) | 2015-05-22 | 2015-05-22 | A kind of content delivery method, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510267665.3A CN104881642B (en) | 2015-05-22 | 2015-05-22 | A kind of content delivery method, device and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104881642A true CN104881642A (en) | 2015-09-02 |
| CN104881642B CN104881642B (en) | 2018-10-26 |
Family
ID=53949130
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510267665.3A Active CN104881642B (en) | 2015-05-22 | 2015-05-22 | A kind of content delivery method, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104881642B (en) |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106412007A (en) * | 2016-08-26 | 2017-02-15 | 珠海格力电器股份有限公司 | Message pushing method and device of intelligent terminal |
| CN106504283A (en) * | 2016-09-26 | 2017-03-15 | 深圳奥比中光科技有限公司 | Information broadcasting method, apparatus and system |
| CN106507145A (en) * | 2016-11-08 | 2017-03-15 | 天脉聚源(北京)传媒科技有限公司 | A kind of live changing method and system based on character features |
| CN106529400A (en) * | 2016-09-26 | 2017-03-22 | 深圳奥比中光科技有限公司 | Mobile terminal and human body monitoring method and device |
| CN106529399A (en) * | 2016-09-26 | 2017-03-22 | 深圳奥比中光科技有限公司 | Human body information acquisition method, device and system |
| WO2017049918A1 (en) * | 2015-09-25 | 2017-03-30 | 杭州海康威视数字技术股份有限公司 | Information publishing terminal and method |
| CN106557937A (en) * | 2015-09-24 | 2017-04-05 | 杭州海康威视数字技术股份有限公司 | Advertisement sending method and device |
| CN106776619A (en) * | 2015-11-20 | 2017-05-31 | 百度在线网络技术（北京）有限公司 | Method and apparatus for determining the attribute information of destination object |
| CN107016321A (en) * | 2016-01-28 | 2017-08-04 | 青岛海尔智能技术研发有限公司 | Commercial cabinet personnel flow detection statistical method |
| CN107463608A (en) * | 2017-06-20 | 2017-12-12 | 上海汇尔通信息技术有限公司 | A kind of information-pushing method and system based on recognition of face |
| CN107483636A (en) * | 2017-09-22 | 2017-12-15 | 百度在线网络技术（北京）有限公司 | Method and apparatus for pushed information |
| CN107807999A (en) * | 2017-11-07 | 2018-03-16 | 杭州盟山网络科技有限公司 | Information-pushing method and device based on camera |
| CN107862241A (en) * | 2017-06-06 | 2018-03-30 | 哈尔滨工业大学深圳研究生院 | A kind of clothes fashion method for digging and visually-perceptible system based on star's identification |
| CN107993599A (en) * | 2017-12-29 | 2018-05-04 | 黄睿 | A kind of clothes recommend method and advertisement machine |
| CN108197971A (en) * | 2017-12-08 | 2018-06-22 | 北京天正聚合科技有限公司 | Information collecting method, information processing method, apparatus and system |
| CN108230033A (en) * | 2018-01-04 | 2018-06-29 | 百度在线网络技术（北京）有限公司 | For the method and apparatus of output information |
| CN109146605A (en) * | 2018-07-06 | 2019-01-04 | 深圳市科脉技术股份有限公司 | Information-pushing method and Related product |
| CN109345953A (en) * | 2018-08-28 | 2019-02-15 | 信利光电股份有限公司 | Advertisement pushing method and system, storage medium and mobile terminal based on advertisement machine |
| CN109753264A (en) * | 2017-11-08 | 2019-05-14 | 阿里巴巴集团控股有限公司 | A kind of task processing method and equipment |
| CN111078014A (en) * | 2019-12-16 | 2020-04-28 | 深圳数拓科技有限公司 | Multidimensional data acquisition and application method and system |
| CN111246296A (en) * | 2020-01-17 | 2020-06-05 | 北京市商汤科技开发有限公司 | Data pushing method and device, electronic equipment and computer storage medium |
| CN111291264A (en) * | 2020-01-23 | 2020-06-16 | 腾讯科技（深圳）有限公司 | Access object prediction method and device based on machine learning and computer equipment |
| WO2020207413A1 (en) * | 2019-04-09 | 2020-10-15 | 华为技术有限公司 | Content pushing method, apparatus, and device |
| CN113421130A (en) * | 2021-07-20 | 2021-09-21 | 重庆京东方智慧科技有限公司 | Advertisement pushing method and device, electronic equipment and readable storage medium |
| CN114157905A (en) * | 2021-11-22 | 2022-03-08 | 深圳康佳电子科技有限公司 | Television sound adjusting method and device based on image recognition and television |
| CN116957680A (en) * | 2023-08-03 | 2023-10-27 | 深圳花旦传媒有限公司 | Advertisement putting effect monitoring system |
| WO2024108606A1 (en) * | 2022-11-26 | 2024-05-30 | 广州视源电子科技股份有限公司 | Person identity recognition method and system, and device and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103514242A (en) * | 2012-12-19 | 2014-01-15 | Tcl集团股份有限公司 | Intelligent interaction method and system for electronic advertising board |
| CN103634680A (en) * | 2013-11-27 | 2014-03-12 | 青岛海信电器股份有限公司 | Smart television play control method and device |
| CN103794159A (en) * | 2012-10-30 | 2014-05-14 | 中国电信股份有限公司 | Method, device and system for playing multimedia advertisements |
| CN104112209A (en) * | 2013-04-16 | 2014-10-22 | 苏州和积信息科技有限公司 | Audience statistical method of display terminal, and audience statistical system of display terminal |
| CN104573619A (en) * | 2014-07-25 | 2015-04-29 | 北京智膜科技有限公司 | Method and system for analyzing big data of intelligent advertisements based on face identification |
-
2015
- 2015-05-22 CN CN201510267665.3A patent/CN104881642B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103794159A (en) * | 2012-10-30 | 2014-05-14 | 中国电信股份有限公司 | Method, device and system for playing multimedia advertisements |
| CN103514242A (en) * | 2012-12-19 | 2014-01-15 | Tcl集团股份有限公司 | Intelligent interaction method and system for electronic advertising board |
| CN104112209A (en) * | 2013-04-16 | 2014-10-22 | 苏州和积信息科技有限公司 | Audience statistical method of display terminal, and audience statistical system of display terminal |
| CN103634680A (en) * | 2013-11-27 | 2014-03-12 | 青岛海信电器股份有限公司 | Smart television play control method and device |
| CN104573619A (en) * | 2014-07-25 | 2015-04-29 | 北京智膜科技有限公司 | Method and system for analyzing big data of intelligent advertisements based on face identification |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106557937A (en) * | 2015-09-24 | 2017-04-05 | 杭州海康威视数字技术股份有限公司 | Advertisement sending method and device |
| CN106557937B (en) * | 2015-09-24 | 2020-02-18 | 杭州海康威视数字技术股份有限公司 | Advertisement pushing method and device |
| CN106557940B (en) * | 2015-09-25 | 2019-09-17 | 杭州海康威视数字技术股份有限公司 | Information release terminal and method |
| WO2017049918A1 (en) * | 2015-09-25 | 2017-03-30 | 杭州海康威视数字技术股份有限公司 | Information publishing terminal and method |
| CN106557940A (en) * | 2015-09-25 | 2017-04-05 | 杭州海康威视数字技术股份有限公司 | Information release terminal and method |
| CN106776619A (en) * | 2015-11-20 | 2017-05-31 | 百度在线网络技术（北京）有限公司 | Method and apparatus for determining the attribute information of destination object |
| CN107016321A (en) * | 2016-01-28 | 2017-08-04 | 青岛海尔智能技术研发有限公司 | Commercial cabinet personnel flow detection statistical method |
| CN106412007A (en) * | 2016-08-26 | 2017-02-15 | 珠海格力电器股份有限公司 | Message pushing method and device of intelligent terminal |
| CN106504283A (en) * | 2016-09-26 | 2017-03-15 | 深圳奥比中光科技有限公司 | Information broadcasting method, apparatus and system |
| CN106529400A (en) * | 2016-09-26 | 2017-03-22 | 深圳奥比中光科技有限公司 | Mobile terminal and human body monitoring method and device |
| CN106529399A (en) * | 2016-09-26 | 2017-03-22 | 深圳奥比中光科技有限公司 | Human body information acquisition method, device and system |
| CN106507145A (en) * | 2016-11-08 | 2017-03-15 | 天脉聚源(北京)传媒科技有限公司 | A kind of live changing method and system based on character features |
| CN107862241B (en) * | 2017-06-06 | 2021-07-20 | 哈尔滨工业大学深圳研究生院 | Clothes fashion mining method based on star recognition and visual perception system |
| CN107862241A (en) * | 2017-06-06 | 2018-03-30 | 哈尔滨工业大学深圳研究生院 | A kind of clothes fashion method for digging and visually-perceptible system based on star's identification |
| CN107463608A (en) * | 2017-06-20 | 2017-12-12 | 上海汇尔通信息技术有限公司 | A kind of information-pushing method and system based on recognition of face |
| CN107483636A (en) * | 2017-09-22 | 2017-12-15 | 百度在线网络技术（北京）有限公司 | Method and apparatus for pushed information |
| CN107807999A (en) * | 2017-11-07 | 2018-03-16 | 杭州盟山网络科技有限公司 | Information-pushing method and device based on camera |
| CN109753264A (en) * | 2017-11-08 | 2019-05-14 | 阿里巴巴集团控股有限公司 | A kind of task processing method and equipment |
| CN108197971A (en) * | 2017-12-08 | 2018-06-22 | 北京天正聚合科技有限公司 | Information collecting method, information processing method, apparatus and system |
| CN107993599A (en) * | 2017-12-29 | 2018-05-04 | 黄睿 | A kind of clothes recommend method and advertisement machine |
| CN108230033A (en) * | 2018-01-04 | 2018-06-29 | 百度在线网络技术（北京）有限公司 | For the method and apparatus of output information |
| CN109146605A (en) * | 2018-07-06 | 2019-01-04 | 深圳市科脉技术股份有限公司 | Information-pushing method and Related product |
| CN109345953A (en) * | 2018-08-28 | 2019-02-15 | 信利光电股份有限公司 | Advertisement pushing method and system, storage medium and mobile terminal based on advertisement machine |
| WO2020207413A1 (en) * | 2019-04-09 | 2020-10-15 | 华为技术有限公司 | Content pushing method, apparatus, and device |
| CN111797249A (en) * | 2019-04-09 | 2020-10-20 | 华为技术有限公司 | Content pushing method, device and equipment |
| EP3944099A4 (en) * | 2019-04-09 | 2022-05-18 | Huawei Technologies Co., Ltd. | Content pushing method, apparatus, and device |
| US11809479B2 (en) | 2019-04-09 | 2023-11-07 | Huawei Technologies Co., Ltd. | Content push method and apparatus, and device |
| CN111078014A (en) * | 2019-12-16 | 2020-04-28 | 深圳数拓科技有限公司 | Multidimensional data acquisition and application method and system |
| CN111078014B (en) * | 2019-12-16 | 2023-11-24 | 深圳数拓科技有限公司 | Multidimensional data acquisition application method and system |
| CN111246296A (en) * | 2020-01-17 | 2020-06-05 | 北京市商汤科技开发有限公司 | Data pushing method and device, electronic equipment and computer storage medium |
| CN111291264A (en) * | 2020-01-23 | 2020-06-16 | 腾讯科技（深圳）有限公司 | Access object prediction method and device based on machine learning and computer equipment |
| CN113421130A (en) * | 2021-07-20 | 2021-09-21 | 重庆京东方智慧科技有限公司 | Advertisement pushing method and device, electronic equipment and readable storage medium |
| CN114157905A (en) * | 2021-11-22 | 2022-03-08 | 深圳康佳电子科技有限公司 | Television sound adjusting method and device based on image recognition and television |
| CN114157905B (en) * | 2021-11-22 | 2023-12-05 | 深圳康佳电子科技有限公司 | Television sound adjusting method and device based on image recognition and television |
| WO2024108606A1 (en) * | 2022-11-26 | 2024-05-30 | 广州视源电子科技股份有限公司 | Person identity recognition method and system, and device and storage medium |
| CN116957680A (en) * | 2023-08-03 | 2023-10-27 | 深圳花旦传媒有限公司 | Advertisement putting effect monitoring system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104881642B (en) | 2018-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104881642A (en) | Method and device for content pushing, and equipment | |
| US7529388B2 (en) | Methods for automatically tracking moving entities entering and exiting a specified region | |
| CN103310444B (en) | A kind of method of the monitoring people counting based on overhead camera head | |
| US20150243031A1 (en) | Method and device for determining at least one object feature of an object comprised in an image | |
| CN107330371A (en) | Acquisition methods, device and the storage device of the countenance of 3D facial models | |
| CN105556508A (en) | Devices, systems and methods of virtualizing a mirror | |
| CN102156985A (en) | Method for counting pedestrians and vehicles based on virtual gate | |
| CN107707839A (en) | Image processing method and device | |
| CN109086724A (en) | A kind of method for detecting human face and storage medium of acceleration | |
| CN106327525B (en) | Cross the border behavior method of real-time for a kind of computer room important place | |
| CN103971103A (en) | People counting system | |
| CN107481067B (en) | Intelligent advertisement system and interaction method thereof | |
| CN107507269A (en) | Personalized three-dimensional model generating method, device and terminal device | |
| CN104182970A (en) | Souvenir photo portrait position recommendation method based on photography composition rule | |
| CN108509857A (en) | Human face in-vivo detection method, electronic equipment and computer program product | |
| CN110032932A (en) | A kind of human posture recognition method based on video processing and decision tree given threshold | |
| CN109711267A (en) | A kind of pedestrian identifies again, pedestrian movement's orbit generation method and device | |
| CN106504283A (en) | Information broadcasting method, apparatus and system | |
| CN106570885A (en) | Background modeling method based on brightness and texture fusion threshold value | |
| JP6080572B2 (en) | Traffic object detection device | |
| JP7099809B2 (en) | Image monitoring system | |
| Karbasi et al. | Real-time hands detection in depth image by using distance with Kinect camera | |
| CN105938603A (en) | Personnel interest degree detection system based on machine vision and personnel interest degree detection method thereof | |
| CN106360941A (en) | Method for locating fingernails and manicure equipment | |
| Rother et al. | What can casual walkers tell us about a 3D scene? |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200818 Address after: 266000, No. 218, Bay Road, Qingdao economic and Technological Development Zone, Shandong Patentee after: Qingdao Hisense Laser Display Co.,Ltd. Address before: 266100 Zhuzhou Road, Laoshan District, Shandong, No. 151, No. Patentee before: HISENSE Co.,Ltd. |
|
| TR01 | Transfer of patent right |