Machine-learning based high efficiency rate control for AV1

Yi CHEN; Yunhao MAO; Shiqi WANG; Xianguo ZHANG; Sam KWONG

doi:10.1109/MIPR54900.2022.00019

Machine-learning based high efficiency rate control for AV1

Yi CHEN, Yunhao MAO, Shiqi WANG, Xianguo ZHANG, Sam KWONG

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

1 Citation (Scopus)

Abstract

Recent years have witnessed the increasing demand of video coding technologies, which have been continuously developed to meet various requirements in video-related applications. Developed by Alliance for Open Media (AOM), the AOMedia Video 1 (AVl) is an open-source and royalty-free standard. Herein, we achieve high efficiency rate control for AVI based on the machine-learning model, which establishes the rate-quantization relationship in a data-driven manner. More specifically, the Supporting Vector Regression (SVR) is used for rate model parameter estimation. The model is trained using sufficient training data, and incorporated in the encoder. Compared to the default rate control scheme in AV 1, experimental results have shown that 2.01% bitrate could be saved with tolerable bitrate error.

Original language	English
Title of host publication	Proceedings of the 5th International Conference on Multimedia Information Processing and Retrieval
Publisher	IEEE
Pages	65-70
Number of pages	6
ISBN (Electronic)	9781665495486
ISBN (Print)	9781665495486
DOIs	https://doi.org/10.1109/MIPR54900.2022.00019
Publication status	Published - Aug 2022
Externally published	Yes
Event	2022 IEEE 5th International Conference on Multimedia Information Processing and Retrieval - , United States Duration: 2 Aug 2022 → 4 Aug 2022

Conference

Conference	2022 IEEE 5th International Conference on Multimedia Information Processing and Retrieval
Country/Territory	United States
Period	2/08/22 → 4/08/22

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

Funding

This work was supported by the Hong Kong Innovation and Technology Commission (InnoHK Project CIMDA), Hong Kong RGC GRF Grant 9042816 (CityU 11209819) and Grant 9042958 (CityU 11203820), as well as the Tencent Rhino-Bird Fund.

Keywords

Quantization parameter (QP)
Rate control
Supporting vector regression
Video coding

Access to Document

10.1109/MIPR54900.2022.00019

Cite this

@inproceedings{b9006b175f3948f8927a4918a625a183,

title = "Machine-learning based high efficiency rate control for AV1",

abstract = "Recent years have witnessed the increasing demand of video coding technologies, which have been continuously developed to meet various requirements in video-related applications. Developed by Alliance for Open Media (AOM), the AOMedia Video 1 (AVl) is an open-source and royalty-free standard. Herein, we achieve high efficiency rate control for AVI based on the machine-learning model, which establishes the rate-quantization relationship in a data-driven manner. More specifically, the Supporting Vector Regression (SVR) is used for rate model parameter estimation. The model is trained using sufficient training data, and incorporated in the encoder. Compared to the default rate control scheme in AV 1, experimental results have shown that 2.01% bitrate could be saved with tolerable bitrate error.",

keywords = "Quantization parameter (QP), Rate control, Supporting vector regression, Video coding",

author = "Yi CHEN and Yunhao MAO and Shiqi WANG and Xianguo ZHANG and Sam KWONG",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE 5th International Conference on Multimedia Information Processing and Retrieval ; Conference date: 02-08-2022 Through 04-08-2022",

year = "2022",

month = aug,

doi = "10.1109/MIPR54900.2022.00019",

language = "English",

isbn = "9781665495486",

pages = "65--70",

booktitle = "Proceedings of the 5th International Conference on Multimedia Information Processing and Retrieval",

publisher = "IEEE",

}

CHEN, Y, MAO, Y, WANG, S, ZHANG, X & KWONG, S 2022, Machine-learning based high efficiency rate control for AV1. in Proceedings of the 5th International Conference on Multimedia Information Processing and Retrieval. IEEE, pp. 65-70, 2022 IEEE 5th International Conference on Multimedia Information Processing and Retrieval, California, United States, 2/08/22. https://doi.org/10.1109/MIPR54900.2022.00019

TY - GEN

T1 - Machine-learning based high efficiency rate control for AV1

AU - CHEN, Yi

AU - MAO, Yunhao

AU - WANG, Shiqi

AU - ZHANG, Xianguo

AU - KWONG, Sam

PY - 2022/8

Y1 - 2022/8

N2 - Recent years have witnessed the increasing demand of video coding technologies, which have been continuously developed to meet various requirements in video-related applications. Developed by Alliance for Open Media (AOM), the AOMedia Video 1 (AVl) is an open-source and royalty-free standard. Herein, we achieve high efficiency rate control for AVI based on the machine-learning model, which establishes the rate-quantization relationship in a data-driven manner. More specifically, the Supporting Vector Regression (SVR) is used for rate model parameter estimation. The model is trained using sufficient training data, and incorporated in the encoder. Compared to the default rate control scheme in AV 1, experimental results have shown that 2.01% bitrate could be saved with tolerable bitrate error.

AB - Recent years have witnessed the increasing demand of video coding technologies, which have been continuously developed to meet various requirements in video-related applications. Developed by Alliance for Open Media (AOM), the AOMedia Video 1 (AVl) is an open-source and royalty-free standard. Herein, we achieve high efficiency rate control for AVI based on the machine-learning model, which establishes the rate-quantization relationship in a data-driven manner. More specifically, the Supporting Vector Regression (SVR) is used for rate model parameter estimation. The model is trained using sufficient training data, and incorporated in the encoder. Compared to the default rate control scheme in AV 1, experimental results have shown that 2.01% bitrate could be saved with tolerable bitrate error.

KW - Quantization parameter (QP)

KW - Rate control

KW - Supporting vector regression

KW - Video coding

UR - http://www.scopus.com/inward/record.url?scp=85139017700&partnerID=8YFLogxK

U2 - 10.1109/MIPR54900.2022.00019

DO - 10.1109/MIPR54900.2022.00019

M3 - Conference paper (refereed)

SN - 9781665495486

SP - 65

EP - 70

BT - Proceedings of the 5th International Conference on Multimedia Information Processing and Retrieval

PB - IEEE

T2 - 2022 IEEE 5th International Conference on Multimedia Information Processing and Retrieval

Y2 - 2 August 2022 through 4 August 2022

ER -

Machine-learning based high efficiency rate control for AV1

Abstract

Conference

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Adaptive Dynamic Range Enhancement Oriented to High Dynamic Display (面向高動態顯示的自適應動態範圍增強)

Cite this

Machine-learning based high efficiency rate control for AV1

Abstract

Conference

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Projects

Adaptive Dynamic Range Enhancement Oriented to High Dynamic Display (面向高動態顯示的自適應動態範圍增強)

Cite this