CN101854554A - Video encoding and decoding system based on image inpainting predication - Google Patents
Video encoding and decoding system based on image inpainting predication Download PDFInfo
- Publication number
- CN101854554A CN101854554A CN 201010199621 CN201010199621A CN101854554A CN 101854554 A CN101854554 A CN 101854554A CN 201010199621 CN201010199621 CN 201010199621 CN 201010199621 A CN201010199621 A CN 201010199621A CN 101854554 A CN101854554 A CN 101854554A
- Authority
- CN
- China
- Prior art keywords
- submodule
- module
- predication
- links
- subelement
- 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 discloses a video encoding and decoding system based on image inpainting predication in the technical field of video processing. The system comprises a macroblock partitioning module, a macroblock predication module, a residual encoding module, a residual decoding module and a macroblock rebuilding module, wherein the macroblock predication module comprises an intra-frame predication submodule, an inter-frame predication submodule, an image impainting predication submodule and a mode selection submodule. In the invention, on the basis of a H.264 mixed video encoding standard system, the image inpainting predication submodule is added, which makes up the deficiency of the traditional video encoding and decoding system; and compared with the traditional video encoding and decoding system H.264, the video encoding efficiency is improved and the system has better adaptability and error resistance.
Description
Technical field
What the present invention relates to is a kind of coding/decoding system of technical field of image processing, specifically is a kind of video coding and decoding system based on image inpainting predication.
Background technology
Video coding and decoding system is used most important for the storage of video and transmission etc.H.264, a new generation's video coding international standard is joint video team (Joint Video Team, JVT) the high compression digital video coding-coding device standard of Ti Chuing that is constituted jointly by ITU-T video coding expert group (VCEG) and ISO/IEC dynamic image expert group (MPEG).
Comprise coding subsystem reconciliation numeral system based on video coding international standard video coding and decoding system H.264, be divided into the macroblock partition module, intra-framed prediction module, inter prediction module, mode selection module and residual coding module in the coding subsystem.Comprise residual error decoding module and macro block rebuilding module in the decoding subsystem.In the coding subsystem, the video file of input is divided into a certain size image block through the macroblock partition module, passes to intra-framed prediction module and inter prediction module, obtains the predicted value of macro block.Through mode selection module, obtain the optimal mode and the residual error data of this macro block.At last by residual coding module output video code stream.In the decoding subsystem, video code flow at first passes through the residual error decoding module, obtains residual error data and macro block mode, obtains output video through the macro block rebuilding module at last.A new generation's video coding international standard gain great majority H.264 come from the reinforcement of details, but the development of technologies of this framework is to a bottleneck phase.In order further to improve the efficient of video coding, the researcher has added the technology in signal processing field, sets up new video coding and decoding system.
Through existing literature search is found that people such as C.B.Zhu were at IEEE International Conference onMultimedia﹠amp in 2008; In " Video coding with spatio-temporaltexture synthesis and edge-based inpainting the video coding of structure repair (synthetic and) " of delivering on the Expo (international conference of IEEE multimedia) based on the space-time texture, a kind of video coding and decoding system based on the structured image recovery technique is proposed, be specially adding video frame structure extraction module and macro block selection module in the coding subsystem, in the decoding subsystem, add macro block and repair module.This system can effectively save the code word of video coding, guarantees the subjective vision effect of decoded video.But this scheme lacks believable assessment objectively, and all processing have only considered the spatial coherence in the video, can only be applied in the I frame coding of video, has certain limitation.
Summary of the invention
The objective of the invention is to overcome above shortcomings in the prior art, a kind of video coding and decoding system based on image inpainting predication is provided.The present invention has realized the raising of code efficiency by add the macroblock prediction pattern based on image repair in H.264 in video coding international standard of new generation, has saving code word overhead, the strong and adaptable advantage of anti-error performance.
The present invention is achieved by the following technical solutions:
The present invention includes: macroblock partition module, macroblock prediction module, residual coding module, residual error decoding module and macro block rebuilding module, wherein: brightness and chrominance information that the macroblock partition module links to each other and transmits macro block with the macroblock prediction module, predicted value and residual error data that the macroblock prediction module links to each other and transmits macro block with the residual coding module, link to each other with the residual error decoding module video code flow of transfer encoding of residual coding module, residual error data and macro block type information that the residual error decoding module links to each other and transmits macro block with the macro block rebuilding module.
Described macroblock prediction module comprises: the infra-frame prediction submodule, the inter prediction submodule, image inpainting predication submodule and model selection submodule, wherein: the infra-frame prediction submodule, brightness and chrominance information that inter prediction submodule and image inpainting predication submodule link to each other and transmit macro block with the macroblock partition module respectively, the model selection submodule respectively with the infra-frame prediction submodule, predicted value and encoder bit rate information that the inter prediction submodule links to each other with the image inpainting predication submodule and transmits macro block, predicted value and residual error data that the model selection submodule links to each other and transmits macro block with the residual coding module.
Described image inpainting predication submodule carries out prediction based on the image repair technology to macro block, comprise: candidate blocks extraction unit, structure conspicuousness metric element and candidate blocks selected cell, wherein: the candidate blocks extraction unit links to each other with the candidate blocks selected cell and transmits all candidate blocks information, structure conspicuousness metric element links to each other with the candidate blocks selected cell and transmits the structural information of current video frame, and the candidate blocks selected cell images that links to each other with the model selection submodule is repaiied the predicted value and the encoder bit rate information of macro block under the prediction.
The architectural feature in described structure conspicuousness metric element prediction current macro zone, comprise: structure extraction subelement, tensor ballot subelement and structure conspicuousness computation subunit, wherein: the structure extraction subelement links to each other with tensor ballot subelement and transmits the structural information of reconstruction regions, tensor ballot subelement links to each other with structure conspicuousness computation subunit and transmits the structural information of the have a few of reconstruction regions to the current macro zone, and structure conspicuousness computation subunit links to each other with the candidate blocks selected cell and transmits the structural information of current video frame.
Described candidate blocks selected cell is that the point of current macro is selected best candidate blocks from all candidate blocks, comprise: potential energy initialization subelement, transmission of messages subelement and confidence level are upgraded subelement, wherein: potential energy initialization subelement links to each other with the transmission of messages subelement and transmits the energy information of point in the current macro, transmission of messages subelement and confidence level are upgraded subelement the link to each other confidence level and the precedence information of point in the transmission current macro, and confidence level is upgraded subelement link to each other with the model selection submodule predicted value and the encoder bit rate information of the macro block of images reparation under predicting.
Described residual coding module is encoded to residual error data and is obtained video code flow, comprise: transformation submodule, quantification submodule and entropy coding submodule, wherein: transformation submodule with quantize link to each other residual error data behind the propagation and transformation of submodule, quantize link to each other with the entropy coding submodule residual error data after the transmission quantification of submodule, the entropy coding submodule video code flow after the transfer encoding that links to each other with the residual error decoding module.
Described residual error decoding module comprises: entropy decoding submodule, inverse quantization submodule and inverse transformation submodule, wherein: the entropy decoding submodule video code flow after the transfer encoding that links to each other with the residual coding module, entropy decoding submodule links to each other with the inverse quantization submodule and transmits decoded residual error data, the inverse quantization submodule residual error data of transmission behind the inverse quantization that link to each other with the inverse transformation submodule, inverse transformation submodule link to each other with the macro block rebuilding module and transmit residual error data after the inverse transformation.
Compared with prior art, the invention has the beneficial effects as follows: can obtain bigger coding gain, improve code efficiency, saved code word overhead, with a wide range of applications, can both handle video I frame coding and P frame coding, also improved anti-error performance simultaneously.
Description of drawings
Fig. 1 is the composition connection diagram of system of the present invention.
Embodiment
Below in conjunction with accompanying drawing embodiments of the invention are elaborated: present embodiment has provided detailed execution mode and process being to implement under the prerequisite with the technical solution of the present invention, but protection scope of the present invention is not limited to following embodiment.
Embodiment
As shown in Figure 1, present embodiment comprises: macroblock partition module, macroblock prediction module, residual coding module, residual error decoding module and macro block rebuilding module, wherein: brightness and chrominance information that the macroblock partition module links to each other and transmits macro block with the macroblock prediction module, predicted value and residual error data that the macroblock prediction module links to each other and transmits macro block with the residual coding module, link to each other with the residual error decoding module video code flow of transfer encoding of residual coding module, residual error data and macro block type information that the residual error decoding module links to each other and transmits macro block with the macro block rebuilding module.
Described macroblock partition module is used for frame of video is divided into the macro block of 16x16 pixel size, obtains macro block data.
Described macroblock prediction module comprises: the infra-frame prediction submodule, the inter prediction submodule, image inpainting predication submodule and model selection submodule, wherein: the infra-frame prediction submodule, brightness and chrominance information that inter prediction submodule and image inpainting predication submodule link to each other and transmit macro block with the macroblock partition module respectively, the model selection submodule respectively with the infra-frame prediction submodule, predicted value and encoder bit rate information that the inter prediction submodule links to each other with the image inpainting predication submodule and transmits macro block, predicted value and residual error data that the model selection submodule links to each other and transmits macro block with the residual coding module.
Described infra-frame prediction submodule is used for handling video encoding standard 9 kinds of 4x4 luminance block predictive modes H.264,4 kinds of 16x16 luminance block predictive modes and 4 kinds of 8x8 chroma block prediction modes.
Described inter prediction submodule is realized motion compensation and the motion estimation process during video encoding standard H.264.
Described image inpainting predication submodule is according to the zone that has rebuild in video former frame and the present frame, obtain the predicted value of current macro, comprise: candidate blocks extraction unit, structure conspicuousness metric element and candidate blocks selected cell, wherein: the candidate blocks extraction unit links to each other with the candidate blocks selected cell and transmits all candidate blocks information, structure conspicuousness metric element links to each other with the candidate blocks selected cell and transmits the structural information of current video frame, and the candidate blocks selected cell images that links to each other with the model selection submodule is repaiied the predicted value and the encoder bit rate information of macro block under the prediction.
Described candidate blocks extraction unit is used for obtaining the video blocks of the big or small preset value T of being as candidate blocks from the reconstruction region of present frame and former frame video.
The architectural feature in described structure conspicuousness metric element prediction current macro zone, comprise: structure extraction subelement, tensor ballot subelement and structure conspicuousness computation subunit, wherein: the structure extraction subelement links to each other with tensor ballot subelement and transmits the structural information of reconstruction regions, tensor ballot subelement links to each other with structure conspicuousness computation subunit and transmits the structural information of the have a few of reconstruction regions to the current macro zone, and structure conspicuousness computation subunit links to each other with the candidate blocks selected cell and transmits the structural information of current video frame.
Described structure extraction subelement is used for the reconstruction region of present frame is extracted structural information.
Described tensor ballot subelement is used for the point in current macro zone is characterized the tensor ballot of architectural feature, and the initiation point of ballot is the present frame point of reconstruction region.
Described structure conspicuousness computation subunit is used for the point in current macro zone is collected voting results and calculated the structure conspicuousness of this point.
Described candidate blocks selected cell is that the point of current macro is selected best candidate blocks from all candidate blocks, comprise: potential energy initialization subelement, transmission of messages subelement and confidence level are upgraded subelement, wherein: potential energy initialization subelement links to each other with the transmission of messages subelement and transmits the energy information of point in the current macro, transmission of messages subelement and confidence level are upgraded subelement the link to each other confidence level and the precedence information of point in the transmission current macro, and confidence level is upgraded subelement link to each other with the model selection submodule predicted value and the encoder bit rate information of the macro block of images reparation under predicting.
Described potential energy initialization subelement is used for current macro is carried out modeling and initialization potential-energy function.
Described transmission of messages subelement is used for pass-along message between the point of current macro model.
Described confidence level is upgraded subelement and is used to the point of current macro model to upgrade confidence information.
Described model selection submodule is chosen best pattern according to the rate-distortion optimization principle in all predictive modes, obtain the macroblock prediction value and the residual error data of corresponding optimal mode.
Described residual coding module is encoded to residual error data and is obtained video code flow, comprise: transformation submodule, quantification submodule and entropy coding submodule, wherein: transformation submodule with quantize link to each other residual error data behind the propagation and transformation of submodule, quantize link to each other with the entropy coding submodule residual error data after the transmission quantification of submodule, the entropy coding submodule video code flow after the transfer encoding that links to each other with the residual error decoding module.
Described transformation submodule is the data that residual error data are transformed to 4x4 integral discrete cosine form.
Quantization step in the described quantification submodule is respectively 24,28,32,35 and 40.
Described entropy coding submodule is to realize by the Variable Length Code based on context-adaptive.
Described residual error decoding module comprises: entropy decoding submodule, inverse quantization submodule and inverse transformation submodule, wherein: the entropy decoding submodule video code flow after the transfer encoding that links to each other with the entropy coding submodule, entropy decoding submodule links to each other with the inverse quantization submodule and transmits decoded residual error data, the inverse quantization submodule residual error data of transmission behind the inverse quantization that link to each other with the inverse transformation submodule, inverse transformation submodule link to each other with the macro block rebuilding module and transmit residual error data after the inverse transformation.
Described entropy decoding submodule is to realize by the variable-length decoding based on context-adaptive.
The inverse quantization step-length of described inverse quantization submodule is respectively 24,28,32,35 and 40.
Described inverse transformation submodule is that the data conversion with 4x4 integral discrete cosine form is a residual error data.
Described macro block rebuilding module is used for reading macro block (mb) type from macro block head data, predicts according to corresponding predictive mode, obtains the predicted value of macro block, and and residual error data addition acquisition decoding video sequence.
The concrete course of work of structure conspicuousness metric element is in the present embodiment:
1) to the R of the reconstruction region Canny operator extraction structural information of present frame, obtain remarkable structure curve,
Described Canny operator is the multistage edge detection algorithm of John F.Canny in exploitation in 1986, follows the tracks of the remarkable structure curve that obtains image by noise reduction, calculating brightness step and edge.
2) each pixel in the reconstruction region of present frame being done tensor according to remarkable structure curve and express, specifically is if some p on remarkable structure curve, is expressed as the second order symmetric tensor according to the direction of structure curve, otherwise is expressed as unit tensor.
3) each pixel in the reconstruction region of present frame is initiated point as ballot, point in the current macro zone carries out the tensor ballot, specifically be that the structure curve direction of initiating to put with ballot is a normal direction, do the circular arc of initiating point and ballot acceptance point through voting, the direction in the sensing center of circle, ballot receiving station place is the direction of ballot.The size of ballot is by length, the curvature of circular arc, and decision.
4) point in the current macro zone carries out synthesis analysis to all ballots that receive, and obtains the structure conspicuousness and estimates, specifically is to all ballot summations, calculates the characteristic value of total ballot matrix, and the structure conspicuousness is by the decision of characteristic value size.The structure conspicuousness is high more, and it is big more to show that this point is in the probability of significant structural region.
The concrete course of work of candidate blocks selected cell is as follows in the present embodiment:
1) the markov random file model is set up in the current macro zone, the data item of potential-energy function and level and smooth item obtain every confidence level and the message function between the consecutive points thus in the definition markov random file.
2) measure opposed belief function by the structure conspicuousness and revise, make the bigger point of structure conspicuousness have higher confidence level.
3) obtain every priority by revised confidence level function calculation, the point of selecting limit priority is to consecutive points pass-along message, and dynamically update confidence level and priority, guarantee that message always spreads out of from the highest point of current priority, constantly carry out this process and all initiate a message and receive the message that all consecutive points pass over to consecutive points up to all points, the transmittance process of message reaches dynamic equilibrium.
4) the some m in the M of current macro zone is carried out candidate blocks and select, obtain the optimal alignment of the candidate blocks in current macro zone.Specifically be according to the confidence level function, select the optimum piece of the candidate blocks of confidence level maximum as this point, the optimum piece of being had a few in the whole macro block zone is formed the optimal alignment of the candidate blocks in this macro block zone.
Implementation result
Adopt respectively present embodiment coding/decoding system and standard H.264 coding/decoding system the 100 frame videos that derive from " highway_cif.yuv " (YUV file of the mobile 4:2:0 form of 352x288) are carried out encoding and decoding:
When code check was 300kbit/s, the PSNR (Y-PSNR) that two kinds of systems obtain was respectively 36.19dB and 36.12dB, is equivalent to the coding gain that the present embodiment system has obtained 0.07dB;
When code check was 200kbit/s, the PSNR that two kinds of systems obtain was respectively 34.44dB and 34.22dB, is equivalent to the coding gain that the present embodiment system has obtained 0.22dB;
When code check is 150kbit/s, the PSNR that two kinds of systems obtain is respectively 33.12dB and 32.63dB, is equivalent to this and has implemented the coding gain that system has obtained 0.49dB;
When code check was 100kbit/s, the PSNR that two kinds of systems obtain was respectively 30.54dB and 29.68dB, is equivalent to the coding gain that the present embodiment system has obtained 0.86dB;
When packet loss was 10%, the PSNR that the decoded video that two kinds of systems obtain compares to original video was 33.80dB and 33.65dB;
When packet loss was 15%, the PSNR that the decoded video that two kinds of systems obtain compares to original video was 33.80dB and 33.64dB.
Therefore, the present embodiment system can obtain bigger coding gain, has stronger anti-error performance.
Claims (10)
1. video coding and decoding system based on image inpainting predication, it is characterized in that, comprise: the macroblock partition module, the macroblock prediction module, the residual coding module, residual error decoding module and macro block rebuilding module, wherein: brightness and chrominance information that the macroblock partition module links to each other and transmits macro block with the macroblock prediction module, predicted value and residual error data that the macroblock prediction module links to each other and transmits macro block with the residual coding module, link to each other with the residual error decoding module video code flow of transfer encoding of residual coding module, residual error data and macro block type information that the residual error decoding module links to each other and transmits macro block with the macro block rebuilding module.
2. the video coding and decoding system based on image inpainting predication according to claim 1, it is characterized in that, described macroblock prediction module comprises: the infra-frame prediction submodule, the inter prediction submodule, image inpainting predication submodule and model selection submodule, wherein: the infra-frame prediction submodule, brightness and chrominance information that inter prediction submodule and image inpainting predication submodule link to each other and transmit macro block with the macroblock partition module respectively, the model selection submodule respectively with the infra-frame prediction submodule, predicted value and encoder bit rate information that the inter prediction submodule links to each other with the image inpainting predication submodule and transmits macro block, predicted value and residual error data that the model selection submodule links to each other and transmits macro block with the residual coding module.
3. the video coding and decoding system based on image inpainting predication according to claim 1, it is characterized in that, described residual coding module is encoded to residual error data and is obtained video code flow, comprise: transformation submodule, quantification submodule and entropy coding submodule, wherein: transformation submodule with quantize link to each other residual error data behind the propagation and transformation of submodule, quantize link to each other with the entropy coding submodule residual error data after the transmission quantification of submodule, the entropy coding submodule video code flow after the transfer encoding that links to each other with the residual error decoding module.
4. the video coding and decoding system based on image inpainting predication according to claim 1, it is characterized in that, described residual error decoding module comprises: entropy decoding submodule, inverse quantization submodule and inverse transformation submodule, wherein: the entropy decoding submodule video code flow after the transfer encoding that links to each other with the residual coding module, entropy decoding submodule links to each other with the inverse quantization submodule and transmits decoded residual error data, the inverse quantization submodule residual error data of transmission behind the inverse quantization that link to each other with the inverse transformation submodule, inverse transformation submodule link to each other with the macro block rebuilding module and transmit residual error data after the inverse transformation.
5. the video coding and decoding system based on image inpainting predication according to claim 2, it is characterized in that, described image inpainting predication submodule carries out prediction based on the image repair technology to macro block, comprise: the candidate blocks extraction unit, structure conspicuousness metric element and candidate blocks selected cell, wherein: the candidate blocks extraction unit links to each other with the candidate blocks selected cell and transmits all candidate blocks information, structure conspicuousness metric element links to each other with the candidate blocks selected cell and transmits the structural information of current video frame, and the candidate blocks selected cell images that links to each other with the model selection submodule is repaiied the predicted value and the encoder bit rate information of macro block under the prediction.
6. the video coding and decoding system based on image inpainting predication according to claim 5, it is characterized in that, described candidate blocks selected cell is that the point of current macro is selected best candidate blocks from all candidate blocks, comprise: potential energy initialization subelement, transmission of messages subelement and confidence level are upgraded subelement, wherein: potential energy initialization subelement links to each other with the transmission of messages subelement and transmits the energy information of point in the current macro, transmission of messages subelement and confidence level are upgraded subelement the link to each other confidence level and the precedence information of point in the transmission current macro, and confidence level is upgraded subelement link to each other with the model selection submodule predicted value and the encoder bit rate information of the macro block of images reparation under predicting.
7. the video coding and decoding system based on image inpainting predication according to claim 5, it is characterized in that, the architectural feature in described structure conspicuousness metric element prediction current macro zone, comprise: the structure extraction subelement, tensor ballot subelement and structure conspicuousness computation subunit, wherein: the structure extraction subelement links to each other with tensor ballot subelement and transmits the structural information of reconstruction regions, tensor ballot subelement links to each other with structure conspicuousness computation subunit and transmits the structural information of the have a few of reconstruction regions to the current macro zone, and structure conspicuousness computation subunit links to each other with the candidate blocks selected cell and transmits the structural information of current video frame.
8. the video coding and decoding system based on image inpainting predication according to claim 3 is characterized in that, the quantization step in the described quantification submodule is respectively 24,28,32,35 and 40.
9. the video coding and decoding system based on image inpainting predication according to claim 3 is characterized in that, described entropy coding submodule is to realize by the Variable Length Code based on context-adaptive.
10. the video coding and decoding system based on image inpainting predication according to claim 4 is characterized in that, the inverse quantization step-length in the described inverse quantization submodule is respectively 24,28,32,35 and 40.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010199621 CN101854554A (en) | 2010-06-13 | 2010-06-13 | Video encoding and decoding system based on image inpainting predication |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010199621 CN101854554A (en) | 2010-06-13 | 2010-06-13 | Video encoding and decoding system based on image inpainting predication |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101854554A true CN101854554A (en) | 2010-10-06 |
Family
ID=42805774
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010199621 Pending CN101854554A (en) | 2010-06-13 | 2010-06-13 | Video encoding and decoding system based on image inpainting predication |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101854554A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013000437A1 (en) * | 2011-06-30 | 2013-01-03 | 华为技术有限公司 | Intra-frame prediction-based decoding method and decoding device |
| CN104365101A (en) * | 2012-04-15 | 2015-02-18 | 三星电子株式会社 | Method and apparatus for determining reference images for inter prediction |
| CN103714420B (en) * | 2013-12-11 | 2017-01-11 | 深圳先进技术研究院 | Object three-dimensional reconstruction method and device |
| CN109687923A (en) * | 2018-07-25 | 2019-04-26 | 中国人民解放军陆军工程大学 | The long-term spectral prediction technique of image reasoning |
| CN112400319A (en) * | 2018-07-18 | 2021-02-23 | 联发科技股份有限公司 | Merging candidates with multiple hypotheses |
| CN113709464A (en) * | 2021-09-01 | 2021-11-26 | 展讯通信(天津)有限公司 | Video coding method and related device |
| US11917185B2 (en) | 2018-07-18 | 2024-02-27 | Hfi Innovation Inc. | Method and apparatus of motion compensation bandwidth reduction for video coding system utilizing multi-hypothesis |
-
2010
- 2010-06-13 CN CN 201010199621 patent/CN101854554A/en active Pending
Non-Patent Citations (4)
| Title |
|---|
| 《IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY》 20071031 Dong Liu et al. Image Compression With Edge-Based Inpainting 1273-1286 1-10 第17卷, 第10期 2 * |
| 《Proceedings of 2006 IEEE International Symposium on Circuits and Systems (ISCAS)》 20061231 Chen Wang et al. Image Compression with Structure-Aware Inpainting 1816-1819 1-10 , 2 * |
| 《Proceedings of 2010 IEEE International Symposium on Circuits and Systems (ISCAS)》 20100602 Yang Xu, Hongkai Xiong Optimized Inpainting-based Macroblock Prediction in Video Compression 3653-3656 1-10 , 2 * |
| 《Proceedings of the 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'03)》 20030620 Jiaya Jia and Chi-Keung Tang Image Repairing: Robust Image Synthesis by Adaptive ND Tensor Voting 5-7 第1卷, 2 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2013000437A1 (en) * | 2011-06-30 | 2013-01-03 | 华为技术有限公司 | Intra-frame prediction-based decoding method and decoding device |
| CN104365101A (en) * | 2012-04-15 | 2015-02-18 | 三星电子株式会社 | Method and apparatus for determining reference images for inter prediction |
| CN104365101B (en) * | 2012-04-15 | 2019-04-19 | 三星电子株式会社 | For determining the method and apparatus of the reference picture for inter-prediction |
| CN103714420B (en) * | 2013-12-11 | 2017-01-11 | 深圳先进技术研究院 | Object three-dimensional reconstruction method and device |
| CN112400319A (en) * | 2018-07-18 | 2021-02-23 | 联发科技股份有限公司 | Merging candidates with multiple hypotheses |
| US11917185B2 (en) | 2018-07-18 | 2024-02-27 | Hfi Innovation Inc. | Method and apparatus of motion compensation bandwidth reduction for video coding system utilizing multi-hypothesis |
| CN112400319B (en) * | 2018-07-18 | 2024-04-26 | 寰发股份有限公司 | Video encoding/decoding method and device |
| CN109687923A (en) * | 2018-07-25 | 2019-04-26 | 中国人民解放军陆军工程大学 | The long-term spectral prediction technique of image reasoning |
| CN109687923B (en) * | 2018-07-25 | 2021-07-06 | 中国人民解放军陆军工程大学 | Long-term frequency spectrum prediction method for image reasoning |
| CN113709464A (en) * | 2021-09-01 | 2021-11-26 | 展讯通信(天津)有限公司 | Video coding method and related device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103248893B (en) | From H.264/AVC standard to code-transferring method and transcoder thereof the fast frame of HEVC standard | |
| EP2214415B1 (en) | A dual prediction video encoding and decoding method and a device | |
| CN101854554A (en) | Video encoding and decoding system based on image inpainting predication | |
| KR100940444B1 (en) | Method of constituting intra prediction mode using spatial edge detection | |
| Ahmad et al. | Energy efficient video compression for wireless sensor networks | |
| CN101835042B (en) | Wyner-Ziv video coding system controlled on the basis of non feedback speed rate and method | |
| EP2202985A1 (en) | An interframe prediction encoding/decoding method and apparatus | |
| CN101267556B (en) | Quick motion estimation method and video coding and decoding method | |
| CN103634606B (en) | Video encoding method and apparatus | |
| CN101014129B (en) | Video data compression method | |
| CN101946516A (en) | The decision of macro block increment quantization parameter fast | |
| CN103181170A (en) | Adaptive scanning of transform coefficients for video coding | |
| CN103561263A (en) | Motion compensation prediction method based on motion vector restraint and weighting motion vector | |
| WO2012167539A1 (en) | Method and device for intra-frame prediction mode processing | |
| CN103327325A (en) | Intra-frame prediction mode rapid self-adaptation selection method based on HEVC standard | |
| CN101022555B (en) | Interframe predictive coding mode quick selecting method | |
| CN102196272B (en) | P frame encoding method and device | |
| CN103929652A (en) | Intra-frame prediction fast mode selecting method based on autoregressive model in video standard | |
| CN103796023A (en) | H.265/HEVC transcoding method and transcoder based on H.264/AVC macro block structure and texture | |
| CN105681797A (en) | Prediction residual based DVC-HEVC (Distributed Video Coding-High Efficiency Video Coding) video transcoding method | |
| CN104079937A (en) | H.264-to-HEVC fast inter-frame transcoding method and device based on motion vector analysis | |
| US20070133689A1 (en) | Low-cost motion estimation apparatus and method thereof | |
| CN110351552B (en) | Fast coding method in video coding | |
| CN110557638A (en) | communication network switching method and system for inspection unmanned aerial vehicle | |
| CN102833536A (en) | Distributed video encoding and decoding method facing to wireless sensor network |
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 |
Open date: 20101006 |