Subblock-Based Motion Derivation and Inter Prediction Refinement in the Versatile Video Coding Standard

被引：36

作者：

Yang, Haitao ^{[1
]}

Chen, Huanbang ^{[1
]}

Chen, Jianle ^{[2
]}

Esenlik, Semih ^{[3
]}

Sethuraman, Sriram ^{[4
]}

Xiu, Xiaoyu ^{[5
]}

Alshina, Elena ^{[3
]}

Luo, Jiancong ^{[6
]}

机构：

[1] Huawei Technol Co Ltd, Shenzhen 518129, Peoples R China

[2] Qualcomm Technol Inc, San Diego, CA 92121 USA

[3] Huawei Technol Duesseldorf GmbH, D-80992 Munich, Germany

[4] Amazon, Prime Video Playback Team, Bengaluru 560052, India

[5] Kwai Inc, San Diego, CA 92122 USA

[6] Apple Inc, Cupertino, CA 95014 USA

来源：

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY | 2021年 / 31卷 / 10期

关键词：

Encoding; Standards; Motion compensation; Tools; Predictive models; Indexes; Bit rate; Versatile video coding (VVC); inter prediction; affine motion compensation (AMC); decoder-side motion vector refinement (DMVR); subblock-based temporal motion vector prediction (SbTMVP); bi-directional optical flow (BDOF); prediction refinement with optical flow (PROF);

D O I：

10.1109/TCSVT.2021.3100744

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Efficient representation and coding of fine-granular motion information is one of the key research areas for exploiting inter-frame correlation in video coding. Representative techniques towards this direction are affine motion compensation (AMC), decoder-side motion vector refinement (DMVR), and subblock-based temporal motion vector prediction (SbTMVP). Fine-granular motion information is derived at subblock level for all the three coding tools. In addition, the obtained inter prediction can be further refined by two optical flow-based coding tools, the bi-directional optical flow (BDOF) for bi-directional inter prediction and the prediction refinement with optical flow (PROF) exclusively used in combination with AMC. The aforementioned five coding tools have been extensively studied and finally adopted in the Versatile Video Coding (VVC) standard. This paper presents technical details of each tool and highlights the design elements with the consideration of typical hardware implementations. Following the common test conditions defined by Joint Video Experts Team (JVET) for the development of VVC, 5.7% bitrate reduction on average is achieved by the five tools. For test sequences characterized by large and complex motion, up to 13.4% bitrate reduction is observed. Additionally, visual quality improvement is demonstrated and analyzed.

引用

页码：3862 / 3877

页数：16

共 49 条

[1]

Alshammari A, 2016, INT CONF COMPUT INFO, P84, DOI 10.1109/ICCITECHN.2016.7860173

[2]

Alshin A., 2010, 2010 28th Picture Coding Symposium (PCS 2010), P422, DOI 10.1109/PCS.2010.5702525

[3]

Alshin A., 2010, JCTVCC204

[4]

Bjontegaard G., 2001, VCEGM33

[5]

Bossen F., 2019, JVETN1010

[6]

Bross B., IEEE T CIRC SYST VID

[7]

Chen C.-C., 2018, JVETK0359

[8]

Chen H., 2018, JVETL0694

[9]

Chen HF, 2015, AER ADV ENG RES, V22, P1, DOI 10.1109/APMC.2015.7411726

[10]

Chen J., 2015, 52 M JUN

← 1 2 3 4 5 →