Video coding optimization algorithm based on rate-distortion characteristic

doi:10.11772/j.issn.1001-9081.2021030398

Journal of Computer Applications ›› 2022, Vol. 42 ›› Issue (3): 946-952.DOI: 10.11772/j.issn.1001-9081.2021030398

• Multimedia computing and computer simulation • Previous Articles Next Articles

Video coding optimization algorithm based on rate-distortion characteristic

Hongwei GUO¹^,², Xiangsuo FAN³^,⁴(), Shuai LIU¹, Xiang WEI¹, Lingli ZHAO¹

^1.School of Engineering，Honghe University，Mengzi Yunnan 661100，China
^2.School of Information and Communication Engineering，University of Electronic Science and Technology of China，Chengdu Sichuan 611731，China
^3.School of Electrical and Information Engineering，Guangxi University of Science and Technology，Liuzhou Guangxi 545006，China
^4.School of Resources and Environment，University of Electronic Science and Technology of China，Chengdu Sichuan 611731，China

Received:2021-03-16 Revised:2021-06-28 Accepted:2021-06-30 Online:2022-04-09 Published:2022-03-10
Contact: Xiangsuo FAN
About author:GUO Hongwei， born in 1980， Ph. D.， professor. His research interests include multimedia communication， video image processing.
LIU Shuai， born in 1979， Ph. D.， professor. His research interests include video image processing.
WEI Xiang， born in 1980， Ph. D.， associate professor. His research interests include video image processing， complex network.
ZHAO Lingli， born in 1979， Ph. D.， professor. Her research interests include video image processing.
Supported by:
National Natural Science Foundation of China(62061015)

基于率失真特性的视频编码优化算法

郭红伟¹^,², 樊香所³^,⁴(), 刘帅¹, 韦相¹, 赵伶俐¹

^1.红河学院工学院, 云南蒙自 661100
^2.电子科技大学信息与通信工程学院, 成都 611731
^3.广西科技大学电气与信息工程学院, 广西柳州 545006
^4.电子科技大学资源与环境学院, 成都 611731

通讯作者: 樊香所
作者简介:郭红伟（1980—），男（彝族），云南金平人，教授，博士，CCF高级会员，主要研究方向：多媒体通信、视频图像处理
刘帅（1979—），男，云南宣威人，教授，博士，主要研究方向：视频图像处理
韦相（1980—），男（壮族），广西百色人，副教授，博士，主要研究方向：视频图像处理、复杂网络
赵伶俐（1979—），女，湖南衡阳人，教授，博士，主要研究方向：视频图像处理。
基金资助:
国家自然科学基金资助项目(62061015)

Abstract

Abstract:

Rate-Distortion （R-D） optimization is a crucial technique in video encoders. However， the widely used independent R-D optimization is far from being global optimal. In order to further improve the compression performance of High Efficiency Video Coding （HEVC）， a two-pass encoding algorithm combined with both R-D dependency and R-D characteristic was proposed. Firstly， the current frame was encoded with the original method in HEVC， and the number of bits consumed by the current frame and the R-D model parameters of each Coding Tree Unit （CTU） were obtained. Then， combined with time domain dependent rate distortion optimization， the optimal Lagrange multiplier and quantization parameter for each CTU were determined according to the information including current frame bit budget and R-D model parameters. Finally， the current frame was re-encoded， where each CTU had different optimization goal according to its Lagrange multiplier. Experimental results show that the proposed algorithm achieves significant rate-distortion performance improvement. Specifically， the proposed algorithm saves 3.5% and 3.8% bitrate at the same coding quality， compared with the original HEVC encoder， under the coding configurations of low-delay B and P frames.

Key words: High Efficiency Video Coding (HEVC), two-pass encoding, rate-distortion optimization, rate-distortion model, adaptive Lagrange multiplier

摘要：

率失真（R-D）优化是视频编码器中一项关键技术，然而当前广泛采用的独立率失真优化远未达到全局最优性能。为了进一步提升高效视频编码（HEVC）的压缩性能，提出了一种结合率失真依赖性和率失真特性的二次编码优化算法。首先，采用原始HEVC的方法对当前帧进行第一次编码，从而得到当前帧消耗的比特数和每个编码树（CTU）单元的率失真模型参数；然后，结合时域依赖率失真优化，根据当前帧比特预算和率失真模型参数计算每个CTU的最佳拉格朗日乘子及量化参数；最后，对当前帧中每个CTU采用不同的优化目标重新编码。实验结果表明，在低延迟B帧和P帧编码配置下，与HEVC基准相比，所提视频编码优化算法在同样编码质量下节省了3.5%和3.8%的码率，获得了显著的率失真性能提升。

关键词: 高效视频编码, 二次编码, 率失真优化, 率失真模型, 自适应拉格朗日乘子

CLC Number:

TN919.81

Hongwei GUO, Xiangsuo FAN, Shuai LIU, Xiang WEI, Lingli ZHAO. Video coding optimization algorithm based on rate-distortion characteristic[J]. Journal of Computer Applications, 2022, 42(3): 946-952.

郭红伟, 樊香所, 刘帅, 韦相, 赵伶俐. 基于率失真特性的视频编码优化算法[J]. 《计算机应用》唯一官方网站, 2022, 42(3): 946-952.

Figures/Tables 7

References 15

1	FLYNN D， MARPE D， NACCARI M， et al. Overview of the range extensions for the HEVC standard： tools， profiles， and performance ［J］. IEEE Transactions on Circuits and Systems for Video Technology， 2016， 26（1）： 4-19. 10.1109/tcsvt.2015.2478707
2	刘妍君，赵志强，刘艳，等.基于质量可伸缩高性能视频编码的帧内快速算法［J］.计算机应用，2018，38（10）：2960-2964.
	LIU Y J， ZHAO Z Q， LIU Y， et al. Fast intra algorithm based on quality scalable high efficiency video coding ［J］. Journal of Computer Applications， 2018， 38（10）： 2960-2964.
3	郭磊，王晓东，徐博文，等.基于HEVC的帧内预测模式决策和编码单元划分快速算法［J］.计算机应用，2018，38（4）：1157-1163.
	GUO L， WANG X D， XU B W， et al. Fast intra mode prediction decision and coding unit partition algorithm based on high efficiency video coding ［J］. Journal of Computer Applications， 2018， 38（4）： 1157-1163.
4	KUANG W， CHAN Y L， TSANG S H， et al. Machine learning-based fast intra mode decision for HEVC screen content coding via decision trees ［J］. IEEE Transactions on Circuits and Systems for Video Technology， 2020， 30（5）： 1481-1496. 10.1109/tcsvt.2019.2903547
5	郭红伟，朱策，刘宇洋. 视频编码率失真优化技术研究综述［J］. 电子学报， 2020， 48（5）： 1018-1029. 10.3969/j.issn.0372-2112.2020.05.024
	GUO H W， ZHU C， LIU Y Y. Overview of rate-distortion optimization for video coding［J］. Acta Electronica Sinica， 2020， 48（5）： 1018-1029. 10.3969/j.issn.0372-2112.2020.05.024
6	ZHAO T S， WANG Z， CHEN C W. Adaptive quantization parameter cascading in HEVC hierarchical coding ［J］. IEEE Transactions on Image Processing， 2016， 25（7）： 2997-3009. 10.1109/tip.2016.2556941
7	XU Y， LI Q， LI X， et al. Efficient QP cascading in H.265/HEVC low-delay prediction ［C］// Proceedings of the 2017 IEEE International Conference on Multimedia & Expo Workshops. Piscataway： IEEE， 2017： 151-156. 10.1109/icmew.2017.8026303
8	YIN H， HUANG X， LI D， et al. Distortion propagation modeling and its applications on frame level quantization control for predictive video coding ［J］. Signal Processing： Image Communication， 2019， 78（2019）： 398-408. 10.1016/j.image.2019.07.016
9	BICHON M， TANOU J L， ROPERT M， et al. Optimal adaptive quantization based on temporal distortion propagation model for HEVC ［J］. IEEE Transactions on Image Processing， 2019， 28（11）： 5419-5434. 10.1109/tip.2019.2919180
10	LI L， LI B， LI H Q， et al. λ domain optimal bit allocation algorithm for high efficiency video coding ［J］. IEEE Transactions on Circuits and Systems for Video Technology， 2018， 28（1）： 130-142. 10.1109/tcsvt.2016.2598672
11	GAO Y B， ZHU C， LI S， et al. Temporal dependent rate-distortion optimization for low-delay hierarchical video coding ［J］. IEEE Transactions on Image Processing， 2017， 26（9）： 4457-4470. 10.1109/tip.2017.2713598
12	XIAO W， LI B， XU J， et al. Weighted rate-distortion optimization for screen content coding ［J］. IEEE Transactions on Circuits and Systems for Video Technology， 2018， 28（2）： 499-512. 10.1109/tcsvt.2016.2612684
13	BOSSEN F. Common test conditions and software reference configurations： JCTVC-L1100 ［R］. Geneva， Switzerland： ITU-T SG16 WP3 and ISO/IEC JTC1/SC29/WG11， 2013：1-4.
14	LI S X， XU M， REN Y， et al. Closed-form optimization on saliency-guided image compression for HEVC-MSP［J］. IEEE Transactions on Multimedia， 2018， 20（1）： 155-170. 10.1109/tmm.2017.2721544
15	LI B， XU J Z， ZHANG D， et al. QP refinement according to Lagrange multiplier for high efficiency video coding ［C］// Proceedings of the 2013 IEEE International Symposium on Circuits and Systems. Piscataway： IEEE， 2013： 477-480. 10.1109/iscas.2013.6571884

序列	QP	LDB配置						LDP配置
		HM16.7		文献［10］算法		本文算法		HM16.7		文献［10］算法		本文算法
		码率/ （kb·s^-1）	PSNR/dB	码率/ （kb·s^-1）	PSNR/dB	码率/ （kb·s^-1）	PSNR/dB	码率/ （kb·s^-1）	PSNR/dB	码率/ （kb·s^-1）	PSNR/dB	码率/ （kb·s^-1）	PSNR/dB
Kimono 1 920×1 080	22	5 200.86	41.68	5 200.88	41.58	5 197.98	41.72	5 478.29	41.55	5 478.55	41.45	5 481.71	41.59
	27	2 400.79	39.66	2 400.93	39.48	2 361.08	39.74	2 474.88	39.54	2 475.10	39.34	2 431.69	39.62
	32	1 150.44	37.13	1 150.50	36.88	1 159.02	37.29	1 194.96	37.01	1 194.78	36.77	1 193.04	37.16
	37	562.03	34.53	562.02	34.29	576.26	34.70	587.33	34.41	587.47	34.17	596.15	34.56
Basketball-Drill 832×480	22	3 668.56	40.26	3 667.10	40.39	3 659.92	40.76	3 917.87	40.16	3 916.13	40.28	3 925.37	40.64
	27	1 720.96	37.02	1 720.93	37.28	1 707.62	37.57	1 803.25	36.95	1 803.24	37.21	1 794.13	37.49
	32	817.74	34.10	817.75	34.32	822.19	34.51	854.50	34.05	854.49	34.29	857.13	34.45
	37	418.13	31.64	418.11	31.84	418.17	31.81	435.22	31.58	435.20	31.85	432.97	31.76
RaceHorses 416×240	22	1 337.63	39.89	1 337.61	39.90	1 330.74	39.93	1 376.74	39.77	1 376.73	39.81	1 368.75	39.81
	27	636.01	35.83	636.03	35.84	636.61	35.95	643.78	35.76	643.84	35.78	644.97	35.87
	32	298.48	32.22	298.51	32.15	304.69	32.39	300.82	32.20	300.84	32.13	306.50	32.37
	37	143.42	29.33	143.43	29.20	146.29	29.44	144.40	29.32	144.42	29.19	147.62	29.43
FourPeople 1 280×720	22	2 225.55	42.52	2 225.63	42.45	2 336.95	42.73	2 433.70	42.38	2 433.81	42.32	2 557.80	42.58
	27	865.47	40.21	865.46	40.09	874.27	40.52	899.70	40.11	899.69	40.01	909.68	40.41
	32	422.87	37.57	422.83	37.51	421.53	37.77	434.44	37.50	434.41	37.47	431.27	37.71
	37	219.11	34.63	219.09	34.88	214.38	34.69	225.20	34.59	225.17	34.86	219.37	34.65

序列	QP	LDB配置						LDP配置
		HM16.7		文献［10］算法		本文算法		HM16.7		文献［10］算法		本文算法
		码率/ （kb·s^-1）	PSNR/dB	码率/ （kb·s^-1）	PSNR/dB	码率/ （kb·s^-1）	PSNR/dB	码率/ （kb·s^-1）	PSNR/dB	码率/ （kb·s^-1）	PSNR/dB	码率/ （kb·s^-1）	PSNR/dB
Kimono 1 920×1 080	22	5 200.86	41.68	5 200.88	41.58	5 197.98	41.72	5 478.29	41.55	5 478.55	41.45	5 481.71	41.59
	27	2 400.79	39.66	2 400.93	39.48	2 361.08	39.74	2 474.88	39.54	2 475.10	39.34	2 431.69	39.62
	32	1 150.44	37.13	1 150.50	36.88	1 159.02	37.29	1 194.96	37.01	1 194.78	36.77	1 193.04	37.16
	37	562.03	34.53	562.02	34.29	576.26	34.70	587.33	34.41	587.47	34.17	596.15	34.56
Basketball-Drill 832×480	22	3 668.56	40.26	3 667.10	40.39	3 659.92	40.76	3 917.87	40.16	3 916.13	40.28	3 925.37	40.64
	27	1 720.96	37.02	1 720.93	37.28	1 707.62	37.57	1 803.25	36.95	1 803.24	37.21	1 794.13	37.49
	32	817.74	34.10	817.75	34.32	822.19	34.51	854.50	34.05	854.49	34.29	857.13	34.45
	37	418.13	31.64	418.11	31.84	418.17	31.81	435.22	31.58	435.20	31.85	432.97	31.76
RaceHorses 416×240	22	1 337.63	39.89	1 337.61	39.90	1 330.74	39.93	1 376.74	39.77	1 376.73	39.81	1 368.75	39.81
	27	636.01	35.83	636.03	35.84	636.61	35.95	643.78	35.76	643.84	35.78	644.97	35.87
	32	298.48	32.22	298.51	32.15	304.69	32.39	300.82	32.20	300.84	32.13	306.50	32.37
	37	143.42	29.33	143.43	29.20	146.29	29.44	144.40	29.32	144.42	29.19	147.62	29.43
FourPeople 1 280×720	22	2 225.55	42.52	2 225.63	42.45	2 336.95	42.73	2 433.70	42.38	2 433.81	42.32	2 557.80	42.58
	27	865.47	40.21	865.46	40.09	874.27	40.52	899.70	40.11	899.69	40.01	909.68	40.41
	32	422.87	37.57	422.83	37.51	421.53	37.77	434.44	37.50	434.41	37.47	431.27	37.71
	37	219.11	34.63	219.09	34.88	214.38	34.69	225.20	34.59	225.17	34.86	219.37	34.65

Class	序列	文献［6］算法		文献［7］算法		文献［10］算法		文献［11］算法		本文算法
Class	序列	LDB	LDP	LDB	LDP	LDB	LDP	LDB	LDP	LDB	LDP
B	BasketballDrive	1.4	1.6	0.5	0.5	3.5	3.2	-2.2	-2.3	-4.8	-4.4
	BQTerrace	2.5	-0.8	0.1	-3.3	7.1	4.5	-0.5	-1.4	0.9	-0.4
	Cactus	-0.9	-0.2	-2.8	-2.3	-1.2	-2.2	-2.3	-2.5	-3.6	-4.0
	Kimono	0.8	0.7	0.4	0.5	6.5	6.6	0.5	-0.1	-3.4	-3.7
	ParkScene	-1.2	-1.1	-1.2	-1.2	2.1	1.7	-2.9	-2.8	-4.2	-4.3
	平均值	0.5	0.0	-0.6	-1.1	3.6	2.8	-1.5	-1.8	-3.0	-3.3
C	BasketballDrill	-2.9	-2.8	-2.3	-2.4	-5.5	-5.9	-6.0	-6.6	-10.7	-10.7
	BQMall	-0.7	-0.6	-0.9	-0.5	-1.7	-1.8	-1.8	-1.6	-3.3	-3.2
	PartyScene	-2.5	-2.4	-1.3	-1.1	-2.4	-2.3	-1.7	-1.7	-3.3	-3.4
	RaceHorses	2.0	1.9	0.3	0.1	-0.2	-1.4	0.9	0.7	-0.5	-0.9
	平均值	-1.0	-1.0	-1.1	-1.0	-2.5	-2.8	-2.2	-2.3	-4.5	-4.5
D	BasketballPass	-0.1	-0.2	-0.8	-0.7	-4.9	-4.8	-2.0	-2.0	-5.7	-5.6
	BlowingBubbles	-1.8	-1.9	-1.0	-0.9	-2.1	-1.7	-1.8	-1.7	-3.1	-3.1
	BQSquare	-2.0	-2.5	-0.4	-0.4	2.3	1.1	0.6	0.2	0.9	-0.3
	RaceHorses	0.9	1.0	0.2	0.3	0.8	0.6	-0.1	0.0	-1.8	-1.9
	平均值	-0.8	-0.9	-0.5	-0.4	-1.0	-1.2	-0.8	-0.9	-2.4	-2.7
E	FourPeople	-4.5	-4.6	-5.6	-5.6	1.7	1.2	-6.9	-7.2	-6.2	-6.6
	Johnny	-3.6	-3.8	-4.4	-4.2	4.9	3.0	-2.7	-3.4	-3.8	-4.9
	KristenAndSara	-4.3	-4.6	-5.5	-5.8	0.0	-0.8	-4.3	-4.8	-3.1	-3.7
	平均值	-4.2	-4.3	-5.2	-5.2	2.2	1.1	-4.7	-5.1	-4.3	-5.1
总平均 BD-rate		-1.1	-1.3	-1.5	-1.7	0.7	0.1	-2.1	-2.3	-3.5	-3.8

Class	序列	文献［6］算法		文献［7］算法		文献［10］算法		文献［11］算法		本文算法
Class	序列	LDB	LDP	LDB	LDP	LDB	LDP	LDB	LDP	LDB	LDP
B	BasketballDrive	1.4	1.6	0.5	0.5	3.5	3.2	-2.2	-2.3	-4.8	-4.4
	BQTerrace	2.5	-0.8	0.1	-3.3	7.1	4.5	-0.5	-1.4	0.9	-0.4
	Cactus	-0.9	-0.2	-2.8	-2.3	-1.2	-2.2	-2.3	-2.5	-3.6	-4.0
	Kimono	0.8	0.7	0.4	0.5	6.5	6.6	0.5	-0.1	-3.4	-3.7
	ParkScene	-1.2	-1.1	-1.2	-1.2	2.1	1.7	-2.9	-2.8	-4.2	-4.3
	平均值	0.5	0.0	-0.6	-1.1	3.6	2.8	-1.5	-1.8	-3.0	-3.3
C	BasketballDrill	-2.9	-2.8	-2.3	-2.4	-5.5	-5.9	-6.0	-6.6	-10.7	-10.7
	BQMall	-0.7	-0.6	-0.9	-0.5	-1.7	-1.8	-1.8	-1.6	-3.3	-3.2
	PartyScene	-2.5	-2.4	-1.3	-1.1	-2.4	-2.3	-1.7	-1.7	-3.3	-3.4
	RaceHorses	2.0	1.9	0.3	0.1	-0.2	-1.4	0.9	0.7	-0.5	-0.9
	平均值	-1.0	-1.0	-1.1	-1.0	-2.5	-2.8	-2.2	-2.3	-4.5	-4.5
D	BasketballPass	-0.1	-0.2	-0.8	-0.7	-4.9	-4.8	-2.0	-2.0	-5.7	-5.6
	BlowingBubbles	-1.8	-1.9	-1.0	-0.9	-2.1	-1.7	-1.8	-1.7	-3.1	-3.1
	BQSquare	-2.0	-2.5	-0.4	-0.4	2.3	1.1	0.6	0.2	0.9	-0.3
	RaceHorses	0.9	1.0	0.2	0.3	0.8	0.6	-0.1	0.0	-1.8	-1.9
	平均值	-0.8	-0.9	-0.5	-0.4	-1.0	-1.2	-0.8	-0.9	-2.4	-2.7
E	FourPeople	-4.5	-4.6	-5.6	-5.6	1.7	1.2	-6.9	-7.2	-6.2	-6.6
	Johnny	-3.6	-3.8	-4.4	-4.2	4.9	3.0	-2.7	-3.4	-3.8	-4.9
	KristenAndSara	-4.3	-4.6	-5.5	-5.8	0.0	-0.8	-4.3	-4.8	-3.1	-3.7
	平均值	-4.2	-4.3	-5.2	-5.2	2.2	1.1	-4.7	-5.1	-4.3	-5.1
总平均 BD-rate		-1.1	-1.3	-1.5	-1.7	0.7	0.1	-2.1	-2.3	-3.5	-3.8

[1]	JIANG Lin, SHI Jiaqi, LI Yuancheng. Parallel design and implementation of synthetic view distortion change algorithm in reconfigurable structure [J]. Journal of Computer Applications, 2021, 41(6): 1734-1740.
[2]	ZHOU Yizhao, WANG Xiaodong, ZHANG Lianjun, LAN Qiongqiong. Selective encryption scheme based on Logistic and Arnold transform in high efficiency video coding [J]. Journal of Computer Applications, 2019, 39(10): 2973-2979.
[3]	GUO Lei, WANG Xiaodong, XU Bowen, WANG Jian. Fast intra mode prediction decision and coding unit partition algorithm based on high efficiency video coding [J]. Journal of Computer Applications, 2018, 38(4): 1157-1163.
[4]	REN Shuai, SUO Li, ZHANG Tao, YANG Tao, MU Dejun. Information hiding algorithm based on 3D high efficiency video coding background [J]. Journal of Computer Applications, 2017, 37(6): 1616-1619.
[5]	CAI Chunting, FENG Gui, WANG Chi, HAN Xue. Robust video watermarking algorithm for HEVC based on intra-frame prediction modes of muli-partitioning [J]. Journal of Computer Applications, 2017, 37(6): 1772-1776.
[6]	CHEN Yongna, ZHOU Yu, WANG Xiaodong, GUO Lei. Video information hiding algorithm based on diamond coding [J]. Journal of Computer Applications, 2017, 37(10): 2806-2812.
[7]	ZHAO Liping, LIN Tao, GONG Xunwei, ZHU Rong. Intra mini-block copy algorithm for screen content coding [J]. Journal of Computer Applications, 2016, 36(7): 1938-1943.
[8]	CHEN Zhixian, WANG Guozhong, ZHAO Haiwu, LI Guoping, TENG Guowei. Improved algorithm for sample adaptive offset filter based on AVS2 [J]. Journal of Computer Applications, 2016, 36(5): 1362-1365.
[9]	LI Peng, PENG Zongju, LI Chihang, CHEN Fen. Fast intra hierarchical algorithm for high efficiency video coding based on texture property and spatial correlation [J]. Journal of Computer Applications, 2016, 36(4): 1085-1091.
[10]	LIU Ying, GAO Xueming, LIM Kengpang. Fast intra-depth decision algorithm for high efficiency video coding [J]. Journal of Computer Applications, 2016, 36(10): 2854-2858.
[11]	ZHANG Minghui, FENG Gui. Robust video watermarking algorithm for high efficiency video coding based on texture direction [J]. Journal of Computer Applications, 2015, 35(9): 2531-2534.
[12]	CHEN Xianyi, ZHAO Liping, CHEN Zhizhong, LIN Tao. 2D intra string copy for screen content coding [J]. Journal of Computer Applications, 2015, 35(9): 2640-2647.
[13]	ZHANG Jun, DONG Lanfang, YU Jiakui. Fast intra prediction algorithm for high efficiency video coding [J]. Journal of Computer Applications, 2015, 35(8): 2327-2331.
[14]	GAO Wenhua, ZHANG Yiyun, WANG Haidong. Error concealment for high efficiency video coding based on block-merging [J]. Journal of Computer Applications, 2015, 35(6): 1744-1748.
[15]	. Rate control optimization algorithm based on Cauchy distribution for H.264/AVC [J]. Journal of Computer Applications, 2010, 30(9): 2444-2448.

Video coding optimization algorithm based on rate-distortion characteristic

基于率失真特性的视频编码优化算法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 7

References 15

Related Articles 15

Recommended Articles

Metrics