基于失真反向传播的时域依赖率失真优化

doi:10.11959/j.issn.1000-436x.2022230

通信学报 ›› 2022, Vol. 43 ›› Issue (12): 222-232.doi: 10.11959/j.issn.1000-436x.2022230

• 学术通信 • 上一篇

基于失真反向传播的时域依赖率失真优化

郭红伟¹^,², 朱策², 杨栩³^,⁴, 罗雷²^,³

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，四川成都 611731
³ 重庆邮电大学通信与信息工程学院，重庆 400065
⁴ 成都师范学院物理与工程技术学院，四川成都 611130

修回日期:2022-10-31 出版日期:2022-12-25 发布日期:2022-12-01
作者简介:郭红伟（1980- ），男，彝族，云南金平人，博士，红河学院教授、电子科技大学在站博士后，主要研究方向为视频编码与通信
朱策（1969- ），男，四川自贡人，博士，电子科技大学教授、博士生导师，主要研究方向为视频编码与通信、视频分析和处理
杨栩（1983- ），男，四川苍溪人，重庆邮电大学博士生、成都师范学院讲师，主要研究方向为全景视频编码与通信
罗雷（1986- ），男，重庆人，博士，重庆邮电大学副教授、电子科技大学在站博士后，主要研究方向为视频编码与通信
基金资助:
国家自然科学基金资助项目(62061015);国家自然科学基金资助项目(62020106011);国家自然科学基金资助项目(U19A2052);云南省基础研究计划基金资助项目(202101AT070051);云南省基础研究计划基金资助项目(202001BA070001-098);云南省中青年学术和技术带头人后备人才基金资助项目(202205AC160055)

Temporal dependent rate-distortion optimization based on distortion backward propagation

Hongwei GUO¹^,², Ce ZHU², Xu YANG³^,⁴, Lei LUO²^,³

¹ School of Engineering, Honghe University, Mengzi 661100, China
² School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China
³ School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China
⁴ School of Physics and Engineering Technology, Chengdu Normal University, Chengdu 611130, China

Revised:2022-10-31 Online:2022-12-25 Published:2022-12-01
Supported by:
The National Natural Science Foundation of China(62061015);The National Natural Science Foundation of China(62020106011);The National Natural Science Foundation of China(U19A2052);The Yunnan Fundamental Research Project(202101AT070051);The Yunnan Fundamental Research Project(202001BA070001-098);The Reserve Talents for Yunnan Young and Middle-aged Academic and Technical Leader(202205AC160055)

摘要/Abstract

摘要：

率失真优化是基于块的混合视频编码器中一项关键技术，然而广泛采用的独立率失真优化远未取得最优编码性能。为提升高效视频编码的率失真性能，提出了一种时域依赖率失真优化算法。首先，采用指数形式率失真函数，推导出时域失真传播因子的计算式；然后，通过预编码获得编码失真和运动补偿预测误差，并采用失真反向传播估算时域失真传播因子；最后，自适应地调节编码树单元的拉格朗日乘子和量化参数，以优化编码比特资源分配。实验结果显示，在低时延编码配置下，相比于高效视频编码的率失真优化算法，所提算法在获得同等重建视频质量的前提下，将所有测试视频码率平均节省4.4%，BasketballDrill序列码率最高节省13.0%。

关键词: 高效视频编码, 率失真优化, 率失真依赖, 失真传播模型, 自适应拉格朗日乘子

Abstract:

Rate-distortion optimization (RDO) is a crucial technique in block based hybrid video encoders.However, the widely used independent RDO is far from obtaining optimal coding performance.To improve the rate-distortion (R-D) performance of high efficiency video coding (HEVC), a temporal dependent RDO algorithm was proposed.Firstly, the formula to calculate temporal distortion propagation factor was derived by using an exponential R-D function.Then, the coding distortion and motion compensation predicted error were obtained by pre-encoding, and the temporal distortion propagation factor was estimated by using distortion backward propagation.Finally, the Lagrange multiplier and quantization parameter of coding tree unit were adaptively adjusted to optimize bit resources allocation.Experimental results show that compared with the original RDO method in HEVC under the low-delay configuration, the proposed algorithm achieves an average 4.4% bit rate reduction for all test sequences, and up to 13.0% bit rate reduction for test sequence BasketballDrill, at the same reconstructed video quality.

Key words: HEVC, rate-distortion optimization, rate-distortion dependency, distortion propagation model, adaptive Lagrange multiplier

中图分类号:

TN919.8

郭红伟, 朱策, 杨栩, 罗雷. 基于失真反向传播的时域依赖率失真优化[J]. 通信学报, 2022, 43(12): 222-232.

Hongwei GUO, Ce ZHU, Xu YANG, Lei LUO. Temporal dependent rate-distortion optimization based on distortion backward propagation[J]. Journal on Communications, 2022, 43(12): 222-232.

图/表 6

图1

图2

表1

不同算法相对于HEVC基准编码器的BD-Rate对比"

Class	序列	文献[9]算法		文献[10]算法		文献[16]算法		文献[18]算法		文献[19]算法		本文算法-调节QP		本文算法		本文算法+调节I帧
Class	序列	LDB	LDP	LDB	LDP	LDB	LDP	LDB	LDP	LDB	LDP	LDB	LDP	LDB	LDP	LDB	LDP
	BasketballDrive	-1.2%	-1.2%	0.5%	0.6%	-3.7%	-4.1%	-3.4%	-3.3%	-1.8%	-1.5%	-0.9%	-0.1%	-2.9%	-2.1%	-2.7%	-1.8%
	BQTerrace	2.7%	3.6%	3.8%	2.9%	0.0%	-1.6%	-0.3%	-2.9%	-1.2%	-1.2%	0.0%	0.0%	-0.5%	-1.1%	-0.5%	-1.3%
B	Cactus	-6.8%	-6.5%	1.4%	2.8%	-3.2%	-3.8%	-3.5%	-3.6%	-2.3%	-1.9%	-2.3%	-0.9%	-4.5%	-3.6%	-7.8%	-6.9%
	Kimono	-2.8%	-3.1%	1.2%	1.2%	-0.9%	-1.8%	-2.6%	-2.9%	-0.8%	-0.9%	-0.4%	-0.5%	-3.1%	-3.4%	-2.0%	-2.3%
	ParkScene	-2.7%	-2.8%	0.7%	1.0%	-4.0%	-4.0%	-4.4%	-4.2%	-2.7%	-2.5%	-3.9%	-3.4%	-6.0%	-5.5%	-6.4%	-6.1%
	BasketballDrill	-5.6%	-5.8%	-4.0%	-4.1%	-8.4%	-9.1%	-9.6%	-9.6%	-4.6%	-4.3%	-8.6%	-7.6%	-13.0%	-12.1%	-15.0%	-14.4%
C	BQMall	-2.3%	-2.4%	-0.5%	-0.5%	-2.8%	-2.8%	-3.9%	-3.8%	-1.5%	-1.2%	-1.6%	-0.9%	-3.4%	-2.8%	-3.5%	-3.0%
	PartyScene	-4.5%	-4.4%	-1.2%	-1.3%	-2.6%	-2.9%	-2.7%	-2.8%	-2.9%	-2.3%	-4.7%	-3.3%	-6.7%	-5.7%	-7.4%	-6.5%
	RaceHorses（WVGA）	0.4%	0.5%	0.6%	0.6%	0.0%	-0.3%	-1.6%	-2.3%	0.7%	0.8%	3.0%	3.4%	2.0%	1.9%	2.4%	2.4%
	BasketballPass	-3.4%	-3.5%	-1.4%	-1.3%	-4.5%	-4.6%	-4.3%	-4.3%	-1.5%	-1.4%	-1.7%	-1.5%	-5.0%	-4.6%	-6.2%	-5.8%
D	BlowingBubble	-3.9%	-3.8%	-2.3%	-2.2%	-2.6%	-2.9%	-3.8%	-3.7%	-1.3%	-1.0%	-1.9%	-1.3%	-3.3%	-3.1%	-3.3%	-3.1%
	BQSquare	-2.5%	-2.6%	-3.4%	-3.9%	0.6%	-0.3%	0.7%	-0.9%	0.0%	-0.2%	0.4%	0.3%	0.2%	-0.9%	-1.5%	-2.5%
	RaceHorses（WQVGA）	-0.5%	-0.5%	0.3%	0.3%	-1.3%	-1.3%	-1.7%	-1.8%	-0.4%	-0.3%	-0.1%	0.3%	-1.9%	-1.7%	-1.4%	-1.3%
	FourPeople	-12.9%	-13.6%	-12.0%	-12.3%	-7.7%	-8.6%	-7.3%	-7.8%	-3.7%	-3.8%	-7.0%	-6.7%	-9.5%	-9.7%	-18.1%	-18.3%
E	Johnny	-9.5%	-10.4%	-8.0%	-8.3%	-2.4%	-3.2%	-5.5%	-6.3%	-1.9%	-1.7%	-3.3%	-3.4%	-4.4%	-4.9%	-12.1%	-13.0%
	KristenAndSara	-14.2%	-14.6%	-10.1%	-10.5%	-4.1%	-4.4%	-5.7%	-6.2%	-2.9%	-3.1%	-5.3%	-5.7%	-6.8%	-7.4%	-17.1%	-18.4%
	BasketballDrillText	-5.7%	-6.0%	-4.5%	-4.7%	-6.7%	-7.2%	-7.8%	-7.9%	-4.1%	-3.8%	-7.2%	-6.3%	-11.0%	-10.4%	-13.5%	-13.2%
F	ChinaSpeed	-7.1%	-7.2%	-6.1%	-6.1%	-2.7%	-2.9%	-3.2%	-3.5%	-1.1%	-1.1%	-1.3%	-1.4%	-3.0%	-3.0%	-8.3%	-8.4%
	SlideEditing	-1.7%	-0.7%	0.1%	-1.0%	-0.6%	-0.2%	-1.0%	-0.6%	0.6%	0.4%	1.8%	2.2%	1.4%	1.4%	0.7%	0.7%
	SlideShow	-12.9%	-13.3%	0.2%	0.3%	-6.4%	-6.6%	-7.1%	-7.4%	-1.4%	-1.5%	-3.2%	-3.1%	-5.9%	-6.6%	-5.0%	-5.4%
	Class B	-2.2%	-2.0%	1.5%	1.7%	-2.4%	-3.0%	-2.8%	-3.4%	-1.8%	-1.6%	-1.5%	-1.0%	-3.4%	-3.1%	-3.9%	-3.7%
	Class C	-3.0%	-3.0%	-1.3%	-1.3%	-3.5%	-3.8%	-4.5%	-4.6%	-2.1%	-1.8%	-3.0%	-2.1%	-5.3%	-4.7%	-5.9%	-5.4%
平均值	Class D	-2.6%	-2.6%	-1.7	-1.8%	-2.0%	-2.3%	-2.3%	-2.7%	-0.8%	-0.7%	-0.8%	-0.6%	-2.5%	-2.6%	-3.1%	-3.2%
	Class E	-12.2%	-12.9%	-10.0%	-10.4%	-4.7%	-5.4%	-6.2%	-6.8%	-2.8%	-2.9%	-5.2%	-5.3%	-6.9%	-7.4%	-15.8%	-16.6%
	Class F	-6.8%	-6.8%	-2.6%	-2.8%	-4.1%	-4.2%	-4.8%	-4.8%	-1.5%	-1.5%	-2.5%	-2.1%	-4.6%	-4.7%	-6.5%	-6.6%
总平均	BD-Rate	-4.9%	-4.9%	-2.2%	-2.3%	-3.2%	-3.6%	-3.9%	-4.3%	-1.8%	-1.7%	-2.4%	-2.0%	-4.4%	-4.3%	-6.4%	-6.4%

表1

图3

表2

表3

参考文献 27

[1]	WIEGAND T , SULLIVAN G J , BJONTEGAARD G ,et al. Overview of the H.264/AVC video coding standard[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2003,13(7): 560-576.
[2]	SULLIVAN G J , OHM J R , HAN W J ,et al. Overview of the high efficiency video coding (HEVC) standard[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2012,22(12): 1649-1668.
[3]	BROSS B , CHEN J L , OHM J R ,et al. Developments in international video coding standardization after AVC,with an overview of versatile video coding (VVC)[J]. Proceedings of the IEEE, 2021,109(9): 1463-1493.
[4]	YOU J , CHOI C , JEONG J . Modified rate distortion optimization using inter-block dependence for H.264/AVC intra coding[J]. IEEE Transactions on Consumer Electronics, 2008,54(3): 1383-1388.
[5]	BICHON M , TANOU J L , ROPERT M ,et al. Inter-block dependencies consideration for intra coding in H.264/AVC and HEVC standards[C]// Proceedings of 2017 IEEE International Conference on Acoustics,Speech and Signal Processing. Piscataway:IEEE Press, 2017: 1537-1541.
[6]	BICHON M , TANOU J L , ROPERT M ,et al. Low complexity joint RDO of prediction units couples for HEVC intra coding[C]// Proceedings of 2018 IEEE International Conference on Acoustics,Speech and Signal Processing. Piscataway:IEEE Press, 2018: 1733-1737.
[7]	XU Y W , LI Q , XIA L ,et al. Efficient QP cascading in H.265/HEVC low-delay prediction[C]// Proceedings of 2017 IEEE International Conference on Multimedia ＆ Expo Workshops. Piscataway:IEEE Press, 2017: 151-156.
[8]	TANG T , LI L , LI J . Improved hierarchical quantisation parameter setting method for screen content coding in high efficiency video coding[J]. IET Image Processing, 2019,13(8): 1382-1390.
[9]	AMER H , YANG E H . Adaptive quantization parameter selection for low-delay HEVC via temporal propagation length estimation[J]. Signal Processing:Image Communication, 2020,84: 1-12.
[10]	GONG Y H , YANG K F , LIU Y ,et al. Quantization parameter cascading for surveillance video coding considering all inter reference frames[J]. IEEE Transactions on Image Processing, 2021,30(6): 5692-5707.
[11]	GONG Y C , WAN S , YANG K F ,et al. Rate-distortion-optimizationbased quantization parameter cascading technique for random-access configuration in H.265/HEVC[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2017,27(6): 1304-1312.
[12]	HE J , YANG E H , YANG F Z ,et al. Adaptive quantization parameter selection for H.265/HEVC by employing inter-frame dependency[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2018,28(12): 3424-3436.
[13]	YIN H B , et al . Distortion propagation modeling and its applications on frame level quantization control for predictive video coding[J]. Signal Processing:Image Communication, 2019,78: 398-408.
[14]	BICHON M , LE TANOU J , ROPERT M ,et al. Optimal adaptive quantization based on temporal distortion propagation model for HEVC[J]. IEEE Transactions on Image Processing:a Publication of the IEEE Signal Processing Society, 2019,28(11): 5419-5434.
[15]	YANG T W , ZHU C , FAN X J ,et al. Source distortion temporal propagation model for motion compensated video coding optimization[C]// Proceedings of 2012 IEEE International Conference on Multimedia and Expo. Piscataway:IEEE Press, 2012: 85-90.
[16]	GAO Y B , ZHU C , LI S ,et al. Temporally dependent rate-distortion optimization for low-delay hierarchical video coding[J]. IEEE Transactions on Image Processing, 2017,26(9): 4457-4470.
[17]	GAO Y B , ZHU C , LI S ,et al. Source distortion temporal propagation analysis for random-access hierarchical video coding optimization[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2019,29(2): 546-559.
[18]	LI S , ZHU C , GAO Y B ,et al. Lagrangian multiplier adaptation for rate-distortion optimization with inter-frame dependency[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2016,26(1): 117-129.
[19]	GUO H W , ZHU C , LIU Y Y ,et al. Temporal dependency based Lagrange multiplier adaptation for HEVC[C]// Proceedings of 2020 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting. Piscataway:IEEE Press, 2020: 1-5.
[20]	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.
[21]	GUO H W , ZHU C , LI S X ,et al. Optimal bit allocation at frame level for rate control in HEVC[J]. IEEE Transactions on Broadcasting, 2019,65(2): 270-281.
[22]	GUO H W , ZHU C , XU M ,et al. Inter-block dependency-based CTU level rate control for HEVC[J]. IEEE Transactions on Broadcasting, 2020,66(1): 113-126.
[23]	郭红伟, 朱策, 周益民 . HEVC 码率控制技术的研究进展[J]. 重庆邮电大学学报(自然科学版), 2018,30(2): 199-209.
	GUO H W , ZHU C , ZHOU Y M . Recent advances on rate control for HEVC[J]. Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition), 2018,30(2): 199-209.
[24]	公衍超, 王玲, 刘颖 ,等. 视频主观观测实验启发的 HEVC 感知帧内码率控制[J]. 通信学报, 2021,42(8): 90-102.
	GONG Y C , WANG L , LIU Y ,et al. HEVC perceptual intra-frame rate control inspired by video subjective observation experiment[J]. Journal on Communications, 2021,42(8): 90-102.
[25]	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.
	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.
[26]	LI B , XU J Z , ZHANG D ,et al. QP refinement according to Lagrange multiplier for high efficiency video coding[C]// Proceedings of 2013 IEEE International Symposium on Circuits and Systems. Piscataway:IEEE Press, 2013: 477-480.
[27]	BOSSEN F . Common test conditions and software reference configurations[R]. 2013.

配置	Class B	Class C	Class D	Class E	Class F	总平均
LDB	24%	24%	22%	20%	22%	23%
LDP	24%	25%	23%	21%	23%	23%

基于失真反向传播的时域依赖率失真优化

Temporal dependent rate-distortion optimization based on distortion backward propagation

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 6

参考文献 27

相关文章 6

Metrics

推荐阅读 0

反向传播长度	Class B	Class C	Class D	Class E	Class F	总平均
1	-3.4%	-3.7%	-2.4%	-4.2%	-2.8%	-3.3%
2	-3.6%	-4.5%	-2.6%	-6.1%	-4.0%	-4.1%
3	-3.1%	-4.7%	-2.6%	-7.4%	-4.7%	-4.3%
4	-2.3%	-4.6%	-2.3%	-8.1%	-5.2%	-4.2%

[1]	张平, 戴金晟, 张育铭, 王思贤, 秦晓琦, 牛凯. 面向语义通信的非线性变换编码[J]. 通信学报, 2023, 44(4): 1-14.
[2]	公衍超, 王玲, 刘颖, 杨楷芳, 林庆帆, 王富平. 视频主观观测实验启发的HEVC感知帧内码率控制[J]. 通信学报, 2021, 42(8): 90-102.
[3]	李跃,杨高波,丁湘陵,朱亚培. 基于学习模型的3D-HEVC提前Merge模式终止算法[J]. 通信学报, 2019, 40(7): 104-113.
[4]	李维,赵凡,张二虎,任鹏. 基于SATD的H.265/HEVC拉格朗日因子选择算法[J]. 通信学报, 2017, 38(6): 68-74.
[5]	郑新资,孙知信. H.264中率失真代价估计模型研究[J]. 通信学报, 2012, 33(6): 143-149.
[6]	徐进,吴志美. 结合人眼视觉特性的率失真优化帧内刷新算法[J]. 通信学报, 2007, 28(3): 35-40.