CN102196291A - Method for coding binocular stereo video - Google Patents
Method for coding binocular stereo video Download PDFInfo
- Publication number
- CN102196291A CN102196291A CN 201110131405 CN201110131405A CN102196291A CN 102196291 A CN102196291 A CN 102196291A CN 201110131405 CN201110131405 CN 201110131405 CN 201110131405 A CN201110131405 A CN 201110131405A CN 102196291 A CN102196291 A CN 102196291A
- Authority
- CN
- China
- Prior art keywords
- frame
- coding
- visual point
- point image
- encoded
- 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
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention relates to a video coding technique and discloses a method for coding a binocular stereo video for solving problems of no flexibility, poor image quality and poor video reconstruction stability in the traditional technology. The technical scheme has the key points as follows: the method for coding the binocular stereo video comprises the following steps of: detecting scene switch by reading a left viewpoint sequence image frame by frame, and analyzing relativity and complexity of a left viewpoint time domain at the same time; and judging the coding type of a left viewpoint image to be coded, regulating an image coding sequence and determining a reference frame selection. After a left viewpoint image coding structure is determined, the coding type of a right viewpoint image is determined and left-right viewpoint images are respectively coded according to a right viewpoint time domain and the relativity of the left-right viewpoints. The invention is suitable for coding the binocular stereo video.
Description
Technical field
The present invention relates to video coding technique, relate to a kind of binocular tri-dimensional frequency coding method specifically.
Background technology
In the video display technology field, main development trend be pursue to natural scene more truly, feature clearly.Nowadays simulated television will progressively be replaced by Digital Television through the development and the application of many decades, and HD digital TV (HDTV) has further improved the authenticity of vision by increasing the resolution that image shows.Present two-dimensional video the limitation aspect the expression natural scene depth information made its can not satisfy people to scene the true and demand of reproduction naturally.
Eyes are when observing object, and object is slightly variant on the image space on the retina of both eyes, is called binocular parallax.Binocular parallax strengthens the three-dimensional perceived effect of people for real world through the fusion of brain visual cortex.Binocular tri-dimensional video utilizes human eye binocular parallax characteristic, use two video cameras to obtain two images of same scenery from two different points of view (the viewpoint spacing generally equals people's binocular spacing), and then the right and left eyes that two images are presented to the people respectively, brain by handle a left side and look and the right view picture between spacing, the depth information of scenery in the perceptual image makes the image of enjoying have strong depth perception and sense true to nature.
At the coding of binocular tri-dimensional video, the MPEG-3DAV of International Standards Organization had once proposed four kinds of stereo scopic video coding schemes based on MPEG-4: first kind, left and right sides viewpoint adopts all independently that MPEG-4 encodes; Second kind, consider left and right sides viewpoint correlation, do not consider the residual error of right viewpoint parallax prediction; The third considers the residual error that left and right sides viewpoint correlation and right viewpoint parallax are predicted; The 4th kind, utilize the time domain classification of MPEG-4 to encode.In addition, also be current research emphasis based on the coding of the single view video and the degree of depth and the coding of the multi-view point video and the degree of depth.
Binocular tri-dimensional video in conventional art coding is general adopt as: IBBPBBP ... IBBPBBP ... fixedly GOP (image sets) structure, coding does not have flexibility, do not fully take into account the scene correlation and the complexity characteristics of left and right sides video sequence, when switching, scene GOP is not repartitioned yet, be easy to generate mosaic phenomenon, cause image quality decrease, the video reconstruction of stability is poor.
Summary of the invention
Technical problem to be solved by this invention is: propose a kind of binocular tri-dimensional frequency coding method, solving encodes in the conventional art does not have a problem of flexibility, poor image quality, video reconstruction of stability difference.
The present invention solves the problems of the technologies described above the technical scheme that is adopted: a kind of binocular tri-dimensional frequency coding method may further comprise the steps:
A. read left visual point image frame by frame to buffering area and carry out the scene detection analysis, judge that current whether occurrence scene switches or present frame is first frame to be encoded, in this way, the type of coding of then setting present frame is the I frame, otherwise enters step b;
B. the type of coding of the current frame to be encoded of left visual point image is judged, to determine that adopting the P frame still is that the B frame is encoded;
C. successively left viewpoint frame to be encoded is carried out reference frame selection, simultaneously according to the coding structure of left visual point image and the correlation between the viewpoint of the left and right sides, determine the type of coding of each frame of right visual point image, respectively left visual point image and right visual point image are encoded according to coded sequence then;
D. add up the current GOP length of left visual point image, and compare, when GOP length equals Imax with the I frame period maximum Imax that sets, and when coded sequence does not also finish, then forcing the type of coding of the current frame to be encoded of setting is the I frame, enters step c, otherwise enters step e;
E. judge whether left and right sides viewpoint sequential coding finishes,, then enter step a, otherwise finish coding if do not finish.
Further, among the step a, judge that the concrete grammar that current whether occurrence scene switches is: setting the threshold value that image scene switches is Ts, and whether the otherness cost of judging present frame and its front and back consecutive frame greater than Ts, in this way, then current the scene switching has taken place.
Further, among the step b, type of coding to the current frame to be encoded of left visual point image is judged, to determine to adopt the P frame still is that the B frame concrete grammar of encoding is: the current frame to be encoded of left visual point image is carried out the P frame respectively, the precoding of B frame, and the prediction residual absolute value summation cost value of image residual error after the Hadmard conversion when calculate adopting the different coding type, if the image prediction residual absolute value summation cost that adopts the precoding of P frame is greater than the image prediction residual absolute value summation cost that adopts the precoding of B frame, the type of coding of then determining current frame to be encoded is the B frame, otherwise its type of coding is the P frame.
Further, among the step c, described coded sequence is: if before the I frame of left visual point image and P frame or P frame and the P frame when having the B frame, then at first first I frame or P frame behind the B frame are encoded, and then the B frame is encoded.
Further, among the step c, successively left viewpoint frame to be encoded is carried out reference frame selection and the concrete grammar of encoding is:
If present frame is the I frame, then directly it to be encoded, the reconstructed value of coding is put into left viewpoint reference frame buffering area and is carried out reference for P frame or B frame;
If present frame is the P frame, then to encode with reference to previous I frame or P frame, the reconstructed value of coding is put into left viewpoint reference buffer district and is carried out reference for P frame or B frame;
If present frame is the B frame, then tentatively determine its front and back reference frame information on time domain according to the reference frame number, obtain final front and back reference frame information by calculating reference frame rate-distortion optimization cost, and then this frame is encoded.
Further, among the step c, the definite of type of coding to right visual point image need consider the coding structure of left visual point image and the correlation of left and right sides visual point image, calculate the rate-distortion optimization cost of right visual point image respectively, to determine the coding structure and the type of right visual point image with reference to left visual point image and right viewpoint time domain reference picture.
Concrete, if greater than the rate-distortion optimization cost with reference to right viewpoint time domain reference picture, the frame current to be encoded in the then right visual point image selects right viewpoint time domain reference picture to encode as the reference frame to right visual point image with reference to left visual point image rate-distortion optimization cost; Otherwise, then select left visual point image to encode as the reference frame.
Further, in the steps d, when left visual point image is encoded I frame or P frame at every turn, the maximum buffer of B frame is filled up.
Further, in the steps d, allow behind each I frame of left visual point image to encode immediately following the B frame.
Further, before step a, also comprise step a0: left and right sides visual point image is carried out color correction, the initialization quantization parameter of the left and right sides of initialization simultaneously visual point image signal, the number of B frame is inserted in threshold value that set scene switches and maximum permission, and the maximum of reference frame number and I frame period is set.
The invention has the beneficial effects as follows: by reading left viewpoint sequence image frame by frame, switching detects to scene, simultaneously correlation and complexity on the left viewpoint time-domain are analyzed, judge the type of coding of left viewpoint image to be encoded, the image encoding order is adjusted, determined its reference frame selection.After having determined left visual point image coding structure,, determine the type of coding of right visual point image according to the correlation between right viewpoint time-domain and left and right sides viewpoint.Adopt method of the present invention, owing to considered the scene and the complexity characteristics of binocular video sequence more, improved the compression efficiency of binocular video coding method, can significantly promote the subjective quality of binocular tri-dimensional video reconstructed image, solve the binocular tri-dimensional video image and switch or the problem that subjective quality descends when violent of moving in scene.
Description of drawings
Fig. 1 is the method flow diagram of the embodiment of the invention.
Embodiment
The invention will be further described below in conjunction with drawings and Examples.
Binocular tri-dimensional video coding in the conventional art does not have flexibility, do not fully take into account the scene correlation and the complexity characteristics of left and right sides video sequence, when scene is switched, GOP is not repartitioned yet, be easy to generate mosaic phenomenon, cause image quality decrease, the video reconstruction of stability is poor; At above-mentioned defective, the present invention proposes a kind of binocular tri-dimensional frequency coding method, by reading left viewpoint sequence image frame by frame, switching detects to scene, simultaneously correlation and complexity on the left viewpoint time-domain are analyzed, judge the type of coding of left viewpoint image to be encoded, the image encoding order is adjusted, determine its reference frame selection.After having determined left visual point image coding structure,, determine the type of coding of right visual point image according to the correlation between right viewpoint time-domain and left and right sides viewpoint.
Adopt method of the present invention, owing to considered the scene and the complexity characteristics of binocular video sequence more, improved the compression efficiency of binocular video coding method, can significantly promote the subjective quality of binocular tri-dimensional video reconstructed image, solve the binocular tri-dimensional video image and switch or the problem that subjective quality descends when violent of moving in scene.
Embodiment:
As shown in Figure 1, the binocular tri-dimensional frequency coding method in this example, adopt following steps to finish:
A0. left and right sides visual point image is carried out color correction, makes left and right sides viewpoint image luminance information basically identical to be encoded, eliminate since left and right sides visual point image collection because the inconsistent error of bringing of brightness; The initialization quantization parameter of the vision signal of initialization simultaneously, the number of B frame is inserted in threshold value that set scene switches and maximum permission, and the maximum of reference frame number and I frame period is set.
A. read left viewpoint video image frame by frame to buffering area and carry out the scene detection analysis, switch or present frame is first frame to be encoded as occurrence scene, the type of coding of setting present frame is the I frame, and begins new GOP coding with this position, otherwise enters step b;
B. the type of coding of the current image to be encoded of left viewpoint is judged, to determine that adopting the P frame still is that the B frame is encoded, concrete grammar is the precoding of left viewpoint present frame being carried out P frame, B frame respectively, the image residual error is through the Hadmard conversion (in the video coding when calculate adopting P frame and B frame type of coding respectively, the Hadmard conversion is used to calculate the SATD value more, i.e. the measurement of video residual signals size) after prediction residual absolute value summation cost value Ci_p and Ci_b.
If 8x8 piece image residual matrix is L8, the Hadmard transformation matrix on 8 rank is H8
Carry out earlier the two-dimentional Hadmard conversion of 8x8 piece in the following method
[L′
8]＝[H
8]×[L
8]×[H
8]
Calculate then [L '
8] the absolute value sum and the normalization of all coefficients, this value is the SATD value of this 8x8 piece.SATD to whole two field picture 8x8 adds up, and can obtain prediction residual absolute value summation cost Ci_p and Ci_b.
If the prediction residual absolute value summation cost value Ci_p that adopts P frame coding is greater than the prediction residual absolute value summation cost value Ci_b that adopts the B frame, judge that then the present frame type of coding is the B frame, otherwise type of coding is the P frame, and calculates the coded sequence of determining frame and determine reference frame selection.
C. according to step a and the definite left visual point image type of coding of step b, successively left viewpoint frame to be encoded is carried out reference frame selection, thereby left and right sides visual point image is encoded, concrete grammar is:
If when having the B frame between left viewpoint I frame and P frame and P frame and the P frame, then earlier first I frame or P frame behind the B frame are encoded, and then the B frame is encoded successively, and calculate the coded sequence of this frame;
Determine that the coding earlier of I frame, P frame and B frame or the order of back coding are, because the continuity of image encoding, the top B frame of each GOP can be encoded with reference to the content of previous GOP during coding, that is to say that can allow to follow closely the B frame behind each I frame encodes;
For realizing GOP coding structure flexibly according to the video sequence characteristics, left viewpoint should be filled up the maximum buffer of B frame, so that be follow-up frame type ready for analysis when each coding I frame or P frame;
According to left viewpoint present image type of coding and the reference frame that the front is determined, under flexible GOP coding structure, it is encoded:
If present frame is the I frame, directly it to be encoded, the reconstructed value of coding is put into left viewpoint reference frame buffering area and is carried out reference for P frame or B frame;
If present frame is the P frame, to encode with reference to previous I frame or P frame, the reconstructed value of coding is put into left viewpoint reference buffer district and is carried out reference for P frame or B frame;
If present frame is the B frame, then tentatively determine its front and back reference frame information on time domain according to the reference frame number, obtain final front and back reference frame information by calculating reference frame rate-distortion optimization cost, and then this frame is encoded.
The type of coding of right visual point image determines to need to consider the coding structure of left visual point image and the correlation of left and right sides visual point image, calculate the rate-distortion optimization cost of right viewpoint image reference left side viewpoint object images to be encoded and right viewpoint time domain reference picture respectively, determine the coding structure and the type of right view picture: if right visual point image with reference to left visual point image rate-distortion optimization cost greater than with reference to the rate-distortion optimization cost of right viewpoint time domain reference picture, the frame current to be encoded in the then right visual point image selects right viewpoint time domain reference picture to encode as the reference frame; Otherwise, then select left visual point image to encode as the reference frame.
D. add up the current GOP length of left viewpoint, and compare, when GOP length equals Imax with the I frame largest interval maximum Imax that sets, and when coded sequence does not also finish, then forcing the type of coding of setting image to be encoded is the I frame, enters step c, otherwise enters step e;
E. judge whether left and right sides viewpoint sequential coding finishes,, then enter step a, otherwise finish coding if do not finish.
The present invention's scheme required for protection comprises but is not limited only to the foregoing description, and those skilled in the art changes the technique effect that reaches same according to the description of the foregoing description to the solution of the present invention easily slightly, and it all belongs to protection scope of the present invention.
Claims (10)
1. binocular tri-dimensional frequency coding method is characterized in that: may further comprise the steps:
A. read left visual point image frame by frame to buffering area and carry out the scene detection analysis, judge that current whether occurrence scene switches or present frame is first frame to be encoded, in this way, the type of coding of then setting present frame is the I frame, otherwise enters step b;
B. the type of coding of the current frame to be encoded of left visual point image is judged, to determine that adopting the P frame still is that the B frame is encoded;
C. successively left viewpoint frame to be encoded is carried out reference frame selection, simultaneously according to the coding structure of left visual point image and the correlation between the viewpoint of the left and right sides, determine the type of coding of each frame of right visual point image, respectively left visual point image and right visual point image are encoded according to coded sequence then;
D. add up the current GOP length of left visual point image, and compare, when GOP length equals Imax with the I frame period maximum Imax that sets, and when coded sequence does not also finish, then forcing the type of coding of the current frame to be encoded of setting is the I frame, enters step c, otherwise enters step e;
E. judge whether left and right sides viewpoint sequential coding finishes,, then enter step a, otherwise finish coding if do not finish.
2. a kind of binocular tri-dimensional frequency coding method as claimed in claim 1, it is characterized in that: among the step a, judge that the current concrete grammar whether occurrence scene switches is: the threshold value of setting the image scene switching is Ts, whether the otherness cost of judging present frame and its front and back consecutive frame is greater than Ts, in this way, then current the scene switching taken place.
3. a kind of binocular tri-dimensional frequency coding method as claimed in claim 2, it is characterized in that: among the step b, type of coding to the current frame to be encoded of left visual point image is judged, to determine to adopt the P frame still is that the B frame concrete grammar of encoding is: the current frame to be encoded of left visual point image is carried out the P frame respectively, the precoding of B frame, and the prediction residual absolute value summation cost value of image residual error after the Hadmard conversion when calculate adopting the different coding type, if the image prediction residual absolute value summation cost that adopts the precoding of P frame is greater than the image prediction residual absolute value summation cost that adopts the precoding of B frame, the type of coding of then determining current frame to be encoded is the B frame, otherwise its type of coding is the P frame.
4. a kind of binocular tri-dimensional frequency coding method as claimed in claim 3, it is characterized in that: among the step c, described coded sequence is: if before the I frame of left visual point image and P frame or P frame and the P frame when having the B frame, then at first first I frame or P frame behind the B frame are encoded, and then the B frame is encoded.
5. a kind of binocular tri-dimensional frequency coding method as claimed in claim 4 is characterized in that: among the step c, successively left viewpoint frame to be encoded is carried out reference frame selection and the concrete grammar of encoding is:
If present frame is the I frame, then directly it to be encoded, the reconstructed value of coding is put into left viewpoint reference frame buffering area and is carried out reference for P frame or B frame;
If present frame is the P frame, then to encode with reference to previous I frame or P frame, the reconstructed value of coding is put into left viewpoint reference buffer district and is carried out reference for P frame or B frame;
If present frame is the B frame, then tentatively determine its front and back reference frame information on time domain according to the reference frame number, obtain final front and back reference frame information by calculating reference frame rate-distortion optimization cost, and then this frame is encoded.
6. a kind of binocular tri-dimensional frequency coding method as claimed in claim 5, it is characterized in that: among the step c, the definite of type of coding to right visual point image need consider the coding structure of left visual point image and the correlation of left and right sides visual point image, calculate the rate-distortion optimization cost of right visual point image respectively, to determine the coding structure and the type of right visual point image with reference to left visual point image and right viewpoint time domain reference picture.
7. a kind of binocular tri-dimensional frequency coding method as claimed in claim 6, it is characterized in that: by calculating the rate-distortion optimization cost of right visual point image with reference to left visual point image and right viewpoint time domain reference picture, with the coding structure of determining right visual point image and the concrete grammar of type be: if right visual point image with reference to left visual point image rate-distortion optimization cost greater than rate-distortion optimization cost with reference to right viewpoint time domain reference picture, the frame current to be encoded in the then right visual point image selects right viewpoint time domain reference picture to encode as the reference frame; Otherwise, then select left visual point image to encode as the reference frame.
8. a kind of binocular tri-dimensional frequency coding method as claimed in claim 7 is characterized in that: in the steps d, when left visual point image is encoded I frame or P frame at every turn, the maximum buffer of B frame is filled up.
9. a kind of binocular tri-dimensional frequency coding method as claimed in claim 8 is characterized in that: in the steps d, allow behind each I frame of left visual point image to encode immediately following the B frame.
10. as any described a kind of binocular tri-dimensional frequency coding method of claim 1 to 9, it is characterized in that: before step a, also comprise step a0: left and right sides visual point image is carried out color correction, the initialization quantization parameter of the left and right sides of initialization simultaneously visual point image signal, the number of B frame is inserted in threshold value that set scene switches and maximum permission, and the maximum of reference frame number and I frame period is set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110131405 CN102196291A (en) | 2011-05-20 | 2011-05-20 | Method for coding binocular stereo video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110131405 CN102196291A (en) | 2011-05-20 | 2011-05-20 | Method for coding binocular stereo video |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102196291A true CN102196291A (en) | 2011-09-21 |
Family
ID=44603565
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110131405 Pending CN102196291A (en) | 2011-05-20 | 2011-05-20 | Method for coding binocular stereo video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102196291A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104363437A (en) * | 2014-11-28 | 2015-02-18 | 广东欧珀移动通信有限公司 | Method and apparatus for recording stereo video |
| CN105898106A (en) * | 2015-12-11 | 2016-08-24 | 乐视云计算有限公司 | Determining method and device of scene switching frame |
| WO2019001072A1 (en) * | 2017-06-30 | 2019-01-03 | 武汉斗鱼网络科技有限公司 | Method for determining type of video frame and electronic device |
| CN113014905A (en) * | 2021-02-24 | 2021-06-22 | 北京金山云网络技术有限公司 | Image frame generation method and device, storage medium and electronic equipment |
| CN113938679A (en) * | 2021-10-09 | 2022-01-14 | 腾讯科技（深圳）有限公司 | Image type determination method, device, equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1450816A (en) * | 2003-04-22 | 2003-10-22 | 上海大学 | Stereo video stream coder/decoder and stereo video coding/decoding system |
| US20070041442A1 (en) * | 2004-02-27 | 2007-02-22 | Novelo Manuel R G | Stereoscopic three dimensional video image digital coding system and method |
| US20070104276A1 (en) * | 2005-11-05 | 2007-05-10 | Samsung Electronics Co., Ltd. | Method and apparatus for encoding multiview video |
| CN101610422A (en) * | 2009-07-21 | 2009-12-23 | 天津大学 | Method for compressing three-dimensional image video sequence |
| CN101621683A (en) * | 2008-07-01 | 2010-01-06 | 邹采荣 | Fast stereo video coding method based on AVS |
| CN101720044A (en) * | 2009-12-10 | 2010-06-02 | 四川长虹电器股份有限公司 | Adaptive frame structure-based AVS coding method |
-
2011
- 2011-05-20 CN CN 201110131405 patent/CN102196291A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1450816A (en) * | 2003-04-22 | 2003-10-22 | 上海大学 | Stereo video stream coder/decoder and stereo video coding/decoding system |
| US20070041442A1 (en) * | 2004-02-27 | 2007-02-22 | Novelo Manuel R G | Stereoscopic three dimensional video image digital coding system and method |
| US20070104276A1 (en) * | 2005-11-05 | 2007-05-10 | Samsung Electronics Co., Ltd. | Method and apparatus for encoding multiview video |
| CN101621683A (en) * | 2008-07-01 | 2010-01-06 | 邹采荣 | Fast stereo video coding method based on AVS |
| CN101610422A (en) * | 2009-07-21 | 2009-12-23 | 天津大学 | Method for compressing three-dimensional image video sequence |
| CN101720044A (en) * | 2009-12-10 | 2010-06-02 | 四川长虹电器股份有限公司 | Adaptive frame structure-based AVS coding method |
Non-Patent Citations (2)
| Title |
|---|
| 《信息技术与标准化》 20100910 曾超 3D产业现状及关键技术 , 第09期 * |
| 《天津大学学报》 20080815 彭强，等 基于有损率失真优化的AVS_M帧内更新算法 第41卷, 第08期 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104363437A (en) * | 2014-11-28 | 2015-02-18 | 广东欧珀移动通信有限公司 | Method and apparatus for recording stereo video |
| CN105898106A (en) * | 2015-12-11 | 2016-08-24 | 乐视云计算有限公司 | Determining method and device of scene switching frame |
| WO2019001072A1 (en) * | 2017-06-30 | 2019-01-03 | 武汉斗鱼网络科技有限公司 | Method for determining type of video frame and electronic device |
| CN113014905A (en) * | 2021-02-24 | 2021-06-22 | 北京金山云网络技术有限公司 | Image frame generation method and device, storage medium and electronic equipment |
| CN113014905B (en) * | 2021-02-24 | 2023-04-14 | 北京金山云网络技术有限公司 | Image frame generation method and device, storage medium and electronic equipment |
| CN113938679A (en) * | 2021-10-09 | 2022-01-14 | 腾讯科技（深圳）有限公司 | Image type determination method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101768857B1 (en) | Generation of high dynamic range images from low dynamic range images in multi-view video coding | |
| KR101345303B1 (en) | Dynamic depth control method or apparatus in stereo-view or multiview sequence images | |
| KR102588146B1 (en) | Multi-view signal codec | |
| RU2518435C2 (en) | Encoder optimisation in stereoscopic video delivery systems | |
| CN101668205B (en) | Self-adapting down-sampling stereo video compressed coding method based on residual error macro block | |
| CN101729892B (en) | Coding method of asymmetric stereoscopic video | |
| CN101466038B (en) | Method for encoding stereo video | |
| CN102186023B (en) | Binocular three-dimensional subtitle processing method | |
| KR100738867B1 (en) | Method for Coding and Inter-view Balanced Disparity Estimation in Multiview Animation Coding/Decoding System | |
| CN103024402B (en) | A kind of free viewpoint video entire frame loss error concealing method | |
| CN101867816A (en) | Stereoscopic video asymmetric compression coding method based on human-eye visual characteristic | |
| JP2014502443A (en) | Depth display map generation | |
| CN103597839A (en) | Video compression implementing resolution tradeoffs and optimization | |
| US20090003455A1 (en) | Method and apparatus for illumination compensation in multi-view video coding | |
| CN104602028B (en) | A kind of three-dimensional video-frequency B frames entire frame loss error concealing method | |
| CN102137258B (en) | Method for controlling three-dimensional video code rates | |
| CN102196291A (en) | Method for coding binocular stereo video | |
| CN100581265C (en) | Processing method for multi-view point video | |
| CN105306954B (en) | A kind of perception stereo scopic video coding based on parallax minimum appreciable error model | |
| EP2756681A1 (en) | Frame-compatible full resolution stereoscopic 3d compression and decompression | |
| CN102710949B (en) | Visual sensation-based stereo video coding method | |
| CN101841726B (en) | Three-dimensional video asymmetrical coding method | |
| CN103220533A (en) | Method for hiding loss errors of three-dimensional video macro blocks | |
| CN103702120B (en) | Subjective distortion estimation method for synthetic viewpoint | |
| WO2017220850A1 (en) | Image encoding method and technical equipment for the same |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110921 |