Gaoxing Chen

Fast intra prediction for HEVC based on pixel gradient statistics and mode refinement

High Efficiency Video Coding (HEVC), a successor to H.264, is the next generation video compression standard. To enhance the coding efficiency of video frames, 35 intra prediction modes adopted in Prediction Unit (PU) from 4×4 to 64×64 of HEVC. However, the improvement is based on the cost of rapid increased complexity performance loss. This paper proposed a Pixel Gradient Statistics (PGS) and Mode Refinement (MR) based fast mode decision algorithm. PGS use pixel gradient information to assist prediction mode selection after Rough Mode Decision (RMD). MR utilizes neighboring mode information to select most probable mode (MPM). Experiment result shows that the proposed method performs about 28% time saving with little degradation (BD-rate increase 0.53% and BDPSNR reduce 0.038) in the coding gain.

Low complexity SAO in HEVC base on class combination, pre-decision and merge separation

High efficiency video coding (HEVC) is the most recent video compression standard that doubled the coding efficiency as compared to the predecessor H.264/AVC. HEVC adopts sample adaptive offset (SAO) right after deblock filtering process. According to the intensity and the edge property, the encoder classifies the reconstructed pixels into different categories, and assigns one offset for each category of pixels. This paper proposes class combination, pre-decision and merge separation to decrease the SAO processing time. Class combination combines the four edge offset (EO) type into one. Pre-decision decreases the band offset (BO) search time from 29 to 4. Merge separation limits the merge range to saving the efficiency loss. Experimental result shows the proposed methods can saving about 38% SAO processing time with little degradation (BD-bitrate increased by 0.16% and ΔPSNR decreased by 0.001) in terms of coding gain.

Fast mode and depth decision HEVC intra prediction based on edge detection and partitioning reconfiguration

High Efficiency Video Coding (HEVC), a successor to H.264, is the next generation video compression standard. To enhance the coding efficiency of video frames, 35 intra prediction modes adopted in Prediction Unit (PU) from 4×4 to 64×64 of HEVC. However the improvement is based on the cost of rapid increased complexity. This paper proposes a fast mode and depth decision algorithm based on edge detection and reconfiguration to alleviate the large computation complexity in intra prediction with the trivial degradation in the accuracy. In mode decision, pixel gradient statistics (PGS) and mode refinement (MR) is proposed. PGS use pixel gradient information to assist prediction mode selection after Rough Mode Decision (RMD). MR utilizes the neighboring mode information to select most probable mode (MPM). In depth decision, unit partitioning reconfiguration (UPR) is proposed. It exchanged the original partitioning order in a more reasonable structure, utilizes detecting the smoothness of coding unit to decide the prediction depth. Smoothness detecting is based on PGS result. Experiment result shows that the proposed method performs about 42.84% time saving with little degradation (BD-rate increase 0.61% and BDPSNR reduce 0.041db) in the coding gain. Comparing with previous work, it achieves more time saving and lesser efficiency loss.

Fast Depth Decision for HEVC Inter Prediction Based on Spatial and Temporal Correlation

High efficiency video coding (HEVC) is a video compression standard that outperforms the predecessor H.264/AVC by doubling the compression efficiency. To enhance the compression accuracy, the partition sizes ranging is from 4x4 to 64x64 in HEVC. However, the manifold partition sizes dramatically increase the encoding complexity. This paper proposes a fast depth decision based on spatial and temporal correlation. Spatial correlation utilize the code tree unit (CTU) Splitting information and temporal correlation utilize the motion vector predictor represented CTU in inter prediction to determine the maximum depth in each CTU. Experimental results show that the proposed method saves about 29.1% of the original processing time with 0.9% of BD-bitrate increase on average.

Content Classification Based Reference Frame Reduction and Machine Learning Based Non-square Block Partition Skipping for Inter Prediction of Screen Content Coding

Screen Content Coding (SCC) is the extension of the latest video compression standard High Efficiency Video Coding (HEVC). SCC is mainly developed for reducing the bit-rate of videos generated from computers. However, under inter configuration, SCC has large complexity which brings heavy burden to encoding. This paper proposes a content classification based reference frame reduction method and a non-square prediction unit (PU) skipping method to accelerate SCC. In reference frame reduction method, according to number of colors, input coding tree unit (CTUs) will be divided into two classes: natural contents and screen contents. For each class, reference frame can be reduced based on different standard. In PU partition skipping method, five features are extracted from a CTU. The classic learning tool SVM is used to classify CTUs, then six non-square PU partition in depth 1, 2, 3 can be skipped. Finally, 40.83% encoding time saving on average is achieved with only 0.71% BD-rate degradation compared with SCC reference software (SCM6.0).

Fast enhancement layer intra coding based on inter-channel correlations and TU depth correlation in SHVC

The scalable extension of High Efficiency Video Coding (SHVC) is now being developed by the Joint Collaborative Team on Video (JCT-VC). In SHVC, the enhancement layer (EL) employs the same coding methods as the base layer (BL) for different color components, namely one luminance (luma) and two chrominance (chroma) color components, which causes heavy computation cost. This paper proposes a fast EL luma and chroma intra coding algorithm for all intra configuration in SHVC. The correlation of prediction modes and transform unit (TU) depth between different color channels from different layers are exploited to simplify the intra prediction process in EL. Experimental results show that approximately 30% and 27% encoding time can be saved in All intra 2x and 1.5x respectively, while the BD-rate degradation are 0.4% and -0.1% compared with SHVC Test Model (SHM) 8.0.

Content based mode and depth skipping with Sharp and Directional Edges for intra prediction in Screen Content Coding

Screen Content Coding (SCC) is the extension of High Efficiency Video Coding (HEVC). Main target of SCC is saving BD-rate for screen videos generated by computers. However encoding time is increased because of new intra modes named Intra Block Copy (IntraBC) and Palette mode to save BD-rate. This paper proposes Sharp Edge Based Classification (SE BC) and Directional Edge Based Classification (DEBC). SEBC classifies Largest Coding Units (LCUs) based on sharp edge features and DEBC classifies LCUs based on directional edge features. Then needless mode and depth are skipped. Experimental results show 10.9% time saving with 2.10% BD-rate increase in average for screen videos at the SCC reference software (SCM3.0 [1]).

Deblocking strength prediction based CTU-level SAO category determination in HEVC encoder

High efficiency video coding (HEVC) is a video compression standard that outperforms the predecessor H.264/AVC by doubling the compression efficiency. To enhance the coding accuracy, HEVC adopts sample adaptive offset (SAO), which reduces the distortion of reconstructed pixels using classification based non-linear filtering. In the traditional coding tree unit (CTU) based VLSI encoder implementation, during the pixel classification stage, SAO cannot use the raw samples in the boundary of the current CTU because these pixels have not been processed by deblocking filter (DF). This paper proposes a category determination algorithm based on estimating the deblocking strengths on CTU boundaries and selectively adopting the promising samples in these areas during SAO classification. Compared with HEVC test mode (HM11.0), experimental results indicate that the proposed method achieves an average 0.15% BD-bitrate reduction (equivalent to 0.0084 dB increases in P-SNR).

Hardware oriented category pre-determination algorithm for SAO in HEVC

High efficiency video coding (HEVC) is a video compression standard that outperforms the predecessor H.264/AVC by doubling the compression efficiency. To enhance the coding accuracy, HEVC adopts sample adaptive offset (SAO) which classifies reconstructed pixels into different categories. During the pixel classification, however, SAO cannot use the raw samples in the current code-tree-block (CTB) boundary 4 × 4-blocks because these pixels need to be processed by deblocking filter first, which poses the problem for hardware design. This paper proposes hardware oriented category pre-determination algorithm which estimates the deblocking strengthens on CTB boundaries and selects the promising samples in these areas during SAO classification. Compared with HEVC test mode (HM12.0), experiment results illustrate that the proposed method achieves the average 0.1% BD-bitrate (BDBR) reduction, or equivalently 0.008 db PSNR increasing.