CN102630033A - Method for converting 2D (Two Dimension) into 3D (Three Dimension) based on dynamic object detection - Google Patents
Method for converting 2D (Two Dimension) into 3D (Three Dimension) based on dynamic object detection Download PDFInfo
- Publication number
- CN102630033A CN102630033A CN2012101005893A CN201210100589A CN102630033A CN 102630033 A CN102630033 A CN 102630033A CN 2012101005893 A CN2012101005893 A CN 2012101005893A CN 201210100589 A CN201210100589 A CN 201210100589A CN 102630033 A CN102630033 A CN 102630033A
- Authority
- CN
- China
- Prior art keywords
- image
- frame
- moving region
- difference
- signal
- 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.)
- Pending
Links
Images
Landscapes
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
Abstract
The invention relates to a method for converting 2D (Two Dimension) into 3D (Three Dimension) based on dynamic object detection. The method comprises the following steps of: firstly detecting an appropriate contour of an input 2D video signal by adopting a difference value method; then carrying out image preprocessing on the input 2D video signal, and obtaining a smooth filtering image by adopting mean value filtering; and carrying out moving target detection by adopting an image sequence frame difference method; obtaining a moving region by combining the contour of the video signal and a moving target; enlarging the left and right eye parallax of the determined moving region to form parallax graphs; and synthesizing the left and right eye parallax graphs, and outputting a 3D signal; and then receiving a signal of the next frame, and repeating the same step. According to the technical scheme, the image moving region is obtained by using the contour detection and moving region detection, the video parallax map is generated in allusion to the moving region, and then the left and right eye parallax graphs are synthesized to obtain the 3D effect.
Description
Technical field
The present invention relates to a kind of flat panel display field, particularly a kind of method that detects 2D commentaries on classics 3D based on dynamic object.
Background technology
Present stage, panel display apparatus has been widely used in the individual consumer's goods such as notebook computer, computer monitor, TV because in light weight, advantages such as volume is little, little power consumption, radiation is low, quality is good are arranged.
Along with the continuous development of science and technology, the 3D technology is more and more ripe, is also more and more accepted by masses.But main video basically all is traditional 2D video at present, and the film source of 3D has only the film and the fragment support of only a few.In order to let the popular 3D effect of better experiencing, 2D changes the 3D technology and is absolutely necessary.
The present invention proposes a kind of 2D changes the 3D method, utilizes profile to detect and the moving region detection, obtains the image motion zone, to the moving region, produces visual parallax figure, and synthetic then right and left eyes view obtains 3D effect.Because human eye is relatively responsive to the image of motion, can be automatic dynamic object during the tracking video playback, so strengthen the right and left eyes parallax to dynamic object, can obtain reasonable 2D changes 3D effect.
Summary of the invention
The objective of the invention is to realize that vision signal 2D changes the 3D method, patent of the present invention realizes through following technical scheme:
At first will import the 2D vision signal adopts differential technique to detect general profile.To import the 2D vision signal then and carry out the image preliminary treatment, adopt mean filter, obtain the smothing filtering image; Carry out moving object detection through image sequence frame difference method then.Profile and moving target through with vision signal combine, and obtain the moving region.The right and left eyes parallax is strengthened in the moving region of confirming, form disparity map.Synthetic right and left eyes view, output 3D signal.And then the signal of reception next frame, repeat identical step.
Through above scheme, utilize profile to detect and the moving region detection, obtain the image motion zone, to the moving region, produce visual parallax figure, synthetic then right and left eyes view obtains 3D effect.
Description of drawings
Fig. 1 changes 3D method method flow chart for 2D.
Embodiment
Do further detailed explanation below in conjunction with accompanying drawing:
At first will import the 2D vision signal adopts differential technique to detect general profile.Be about to adjacent two pixel value differences of image, passing threshold detects then.If promptly adjacent two pixel difference are then thought the edge greater than 50% of high luminance values, keep this pixel; If adjacent two pixel difference less than 50% of high luminance values, think that then this place is not the edge, gives up this pixel.
To import the 2D vision signal then and carry out the image preliminary treatment, adopt mean filter, obtain the smothing filtering image.Mean filter is exactly in image processing, and 8 pixels around all using each pixel are done equal Value Operations.As shown in the table:
| 5 | 3 | 6 |
| 2 | 1 | 9 |
| 8 | 4 | 7 |
Very obvious, the gray value of the area pixel of this 3x3 is respectively 5,3,6,2,1,9,8,4,7, and the value after 1 so middle this pixel is filtered is exactly the mean value of these values, just (5+3+6+2+1+9+8+4+7)/9=5.Consider the realization effect of method and the complexity of method, can carry out 1-3 time mean filter usually, just can obtain reasonable effect.
Carry out moving object detection through image sequence frame difference method then; It is poor that the promptly first difference of calculating present frame and former frame obtains forward frame; It is poor to frame that the difference of calculating next frame and present frame then obtains the back, and forward frame difference of asking and the common factor of back to the frame difference obtain the coarse movement zone of moving target.
Profile and moving target through with vision signal combine, and obtain the moving region.The right and left eyes parallax is strengthened in the moving region of confirming, form disparity map.Original video signal is regarded as left eye or right-eye view, parallel the moving in motion parts zone obtained image and be regarded as right eye or left-eye view, the white space of translation is stretched by the zone, next door and fills up, and just can obtain two width of cloth images, and this two width of cloth image is exactly so-called disparity map.Synthetic right and left eyes view promptly obtains the 3D signal.And then the signal of reception next frame, repeat identical step.
Claims (4)
1. a method of changeing 3D based on dynamic object detection 2D is characterized in that, comprises the steps:
1) will import the 2D vision signal adopts differential technique to detect general profile;
2) will import the 2D vision signal and carry out the image preliminary treatment, adopt mean filter, obtain the smothing filtering image; Carry out moving object detection through image sequence frame difference method then;
3) combine through profile and moving target, obtain the moving region vision signal;
4) the right and left eyes parallax is strengthened in the moving region of confirming, form disparity map;
5) synthetic right and left eyes view, output 3D signal;
6) receive the next frame signal, repeat (1).
2. method according to claim 1 is characterized in that: the differential technique described in the step 1), adjacent two pixel value differences of computed image; If greater than 50% of high luminance values, then think the edge, keep this pixel; As less than 50% of high luminance values, then give up this pixel.
3. method according to claim 1; It is characterized in that: step 2) middle image preliminary treatment, adopt mean filter, be to adopt the 3*3 matrix that image is carried out mean filter; Promptly in each 3*3 matrix of image, the numerical value of its central point is averaged for 9 pixel number additions of this matrix.
4. method according to claim 1; It is characterized in that: in step 2) in adopt image sequence frame difference method to carry out moving object detection; It is poor that the promptly first difference of calculating present frame and former frame obtains forward frame; It is poor to frame that the difference of calculating next frame and present frame then obtains the back, and forward frame difference of asking and the common factor of back to the frame difference obtain the coarse movement zone of moving target.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012101005893A CN102630033A (en) | 2012-03-31 | 2012-03-31 | Method for converting 2D (Two Dimension) into 3D (Three Dimension) based on dynamic object detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012101005893A CN102630033A (en) | 2012-03-31 | 2012-03-31 | Method for converting 2D (Two Dimension) into 3D (Three Dimension) based on dynamic object detection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102630033A true CN102630033A (en) | 2012-08-08 |
Family
ID=46588174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012101005893A Pending CN102630033A (en) | 2012-03-31 | 2012-03-31 | Method for converting 2D (Two Dimension) into 3D (Three Dimension) based on dynamic object detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102630033A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107767412A (en) * | 2017-09-11 | 2018-03-06 | 西安中兴新软件有限责任公司 | A kind of image processing method and device |
| CN109688397A (en) * | 2017-10-18 | 2019-04-26 | 上海质尊文化传媒发展有限公司 | A kind of 2D switchs to the method for 3D video |
| CN110460892A (en) * | 2018-05-08 | 2019-11-15 | 日本聚逸株式会社 | Dynamic image dissemination system, dynamic image distribution method and recording medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101030259A (en) * | 2006-02-28 | 2007-09-05 | 沈阳东软软件股份有限公司 | SVM classifier, method and apparatus for discriminating vehicle image therewith |
| CN101795400A (en) * | 2010-03-16 | 2010-08-04 | 上海复控华龙微系统技术有限公司 | Method for actively tracking and monitoring infants and realization system thereof |
| CN101945301A (en) * | 2010-09-28 | 2011-01-12 | 彩虹集团公司 | Method for conversing 2D to 3D of character scene |
| CN102244803A (en) * | 2011-07-19 | 2011-11-16 | 彩虹集团公司 | Device with 3D display function and driving method thereof |
| CN102254147A (en) * | 2011-04-18 | 2011-11-23 | 哈尔滨工业大学 | Method for identifying long-distance space motion target based on stellar map matching |
-
2012
- 2012-03-31 CN CN2012101005893A patent/CN102630033A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101030259A (en) * | 2006-02-28 | 2007-09-05 | 沈阳东软软件股份有限公司 | SVM classifier, method and apparatus for discriminating vehicle image therewith |
| CN101795400A (en) * | 2010-03-16 | 2010-08-04 | 上海复控华龙微系统技术有限公司 | Method for actively tracking and monitoring infants and realization system thereof |
| CN101945301A (en) * | 2010-09-28 | 2011-01-12 | 彩虹集团公司 | Method for conversing 2D to 3D of character scene |
| CN102254147A (en) * | 2011-04-18 | 2011-11-23 | 哈尔滨工业大学 | Method for identifying long-distance space motion target based on stellar map matching |
| CN102244803A (en) * | 2011-07-19 | 2011-11-16 | 彩虹集团公司 | Device with 3D display function and driving method thereof |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107767412A (en) * | 2017-09-11 | 2018-03-06 | 西安中兴新软件有限责任公司 | A kind of image processing method and device |
| CN109688397A (en) * | 2017-10-18 | 2019-04-26 | 上海质尊文化传媒发展有限公司 | A kind of 2D switchs to the method for 3D video |
| CN109688397B (en) * | 2017-10-18 | 2021-10-22 | 上海质尊文化传媒发展有限公司 | Method for converting 2D (two-dimensional) video into 3D video |
| CN110460892A (en) * | 2018-05-08 | 2019-11-15 | 日本聚逸株式会社 | Dynamic image dissemination system, dynamic image distribution method and recording medium |
| CN110460892B (en) * | 2018-05-08 | 2022-06-14 | 日本聚逸株式会社 | Moving image distribution system, moving image distribution method, and recording medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8644596B1 (en) | Conversion of monoscopic visual content using image-depth database | |
| CN101516040B (en) | Video matching method, device and system | |
| TW201243763A (en) | Method for 3D video content generation | |
| CN104539935B (en) | The adjusting method and adjusting means of brightness of image, display device | |
| CN102665086B (en) | Method for obtaining parallax by using region-based local stereo matching | |
| CN102883175B (en) | Methods for extracting depth map, judging video scene change and optimizing edge of depth map | |
| US20120293489A1 (en) | Nonlinear depth remapping system and method thereof | |
| WO2011033673A1 (en) | Image processing apparatus | |
| CN102523464A (en) | Depth image estimating method of binocular stereo video | |
| US10154242B1 (en) | Conversion of 2D image to 3D video | |
| US20120163701A1 (en) | Image processing device, image processing method, and program | |
| EP2560398A2 (en) | Method and apparatus for correcting errors in stereo images | |
| CN101282492A (en) | Method for regulating display depth of three-dimensional image | |
| CN111598932A (en) | Generating a depth map for an input image using an example approximate depth map associated with an example similar image | |
| TW201029443A (en) | Method and device for generating a depth map | |
| CN103053165B (en) | Method for converting 2D into 3D based on image motion information | |
| JP2011223566A (en) | Image converting device and three-dimensional image display device including the same | |
| CN101287142A (en) | Method for converting flat video to tridimensional video based on bidirectional tracing and characteristic points correction | |
| US20180270400A1 (en) | Liquid crystal display device and image processing method for same | |
| US20150195510A1 (en) | Method of integrating binocular stereo video scenes with maintaining time consistency | |
| Didyk et al. | Apparent stereo: The cornsweet illusion can enhance perceived depth | |
| CN102447925A (en) | Virtual viewpoint image synthesis method and device | |
| CN102026012B (en) | Generation method and device of depth map through three-dimensional conversion to planar video | |
| CN102630033A (en) | Method for converting 2D (Two Dimension) into 3D (Three Dimension) based on dynamic object detection | |
| CN102137267A (en) | Algorithm for transforming two-dimensional (2D) character scene into three-dimensional (3D) character scene |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120808 |