Speedup evaluation of HEVC parallel video coding using Tiles

This paper presents an extensive evaluation of the HEVC parallel video coding when using Tiles. The evaluation consists on finding the tiling pattern that yields the maximum possible speedup for a set of video sequences considering several encoding parameters, measuring the coding efficiency variation of using such tiling pattern instead of the uniform tiling pattern, and calculating how far from the uniform tiling the maximum speedup tiling pattern is. To perform these evaluations, a different number of Tiles with different tiling patterns are applied; apart from that, different encoding profiles are employed. The results show that the speedup yielded by the uniform tiling is highly dependent on the video sequence and employed encoding profile. When encoding a set of video sequences with the same encoding parameters, the greater speedup may be up to 25% higher than the minor speedup, whereas when encoding the same video sequence with different encoding profiles, the greater speedup may be up to 21% higher than the minor speedup. When applying the maximum speedup tiling pattern to an encoding, distinct speedup gains may be achieved. While for some video sequences the maximum possible speedup equals the speedup yielded by the uniform tiling pattern, for others, changing from the uniform tiling to a better one may result in more than 40% of speedup gain. The results also show that when changing from the uniform tiling pattern to one that results in the maximum possible speedup, the coding efficiency variation is negligible; therefore, it is rewarding to seek better tiling patterns.

[1]  Ben H. H. Juurlink,et al.  Parallel Scalability and Efficiency of HEVC Parallelization Approaches , 2012, IEEE Transactions on Circuits and Systems for Video Technology.

[2]  Sergio Bampi,et al.  Adaptive content-based Tile partitioning algorithm for the HEVC standard , 2013, 2013 Picture Coding Symposium (PCS).

[3]  Muhammad Usman Karim Khan,et al.  Power efficient and workload balanced tiling for parallelized high efficiency video coding , 2014, 2014 IEEE International Conference on Image Processing (ICIP).

[4]  Guilherme Corrêa,et al.  Performance and Computational Complexity Assessment of High-Efficiency Video Encoders , 2012, IEEE Transactions on Circuits and Systems for Video Technology.

[5]  Yong-Jo Ahn,et al.  Complexity model based load-balancing algorithm for parallel tools of HEVC , 2013, 2013 Visual Communications and Image Processing (VCIP).

[6]  Yongdong Zhang,et al.  High Efficiency Video Coding: High Efficiency Video Coding , 2014 .

[7]  K. Rijkse,et al.  H.263: video coding for low-bit-rate communication , 1996, IEEE Commun. Mag..

[8]  G. Bjontegaard,et al.  Calculation of Average PSNR Differences between RD-curves , 2001 .

[9]  F. Bossen,et al.  Common test conditions and software reference configurations , 2010 .

[10]  Gary J. Sullivan,et al.  Comparison of the Coding Efficiency of Video Coding Standards—Including High Efficiency Video Coding (HEVC) , 2012, IEEE Transactions on Circuits and Systems for Video Technology.

[11]  Minhua Zhou,et al.  An Overview of Tiles in HEVC , 2013, IEEE Journal of Selected Topics in Signal Processing.

[12]  Bruno Zatt,et al.  Speedup-aware history-based tiling algorithm for the HEVC standard , 2016, 2016 IEEE International Conference on Image Processing (ICIP).

[13]  Guilherme Corrêa,et al.  Memory-Aware Tiles Workload Balance through Machine-Learnt Complexity Reduction for HEVC , 2018, 2018 25th IEEE International Conference on Electronics, Circuits and Systems (ICECS).

[14]  Jiangtao Wen,et al.  Wavefront Parallel Processing for AV1 Encoder , 2018, 2018 Picture Coding Symposium (PCS).

[15]  Qionghai Dai,et al.  Clustering-Based Content Adaptive Tiles Under On-chip Memory Constraints , 2016, IEEE Transactions on Multimedia.