Rate Adaptation for Learned Two-layer B-frame Coding without Signaling Motion Information

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Authors

Research Organisations

External Research Organisations

  • National Yang Ming Chiao Tung University (NSTC)
View graph of relations

Details

Original languageEnglish
Title of host publication2023 IEEE International Conference on Visual Communications and Image Processing, VCIP 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (electronic)9798350359855
ISBN (print)979-8-3503-5986-2
Publication statusPublished - 2023
Event2023 IEEE International Conference on Visual Communications and Image Processing, VCIP 2023 - Jeju, Korea, Republic of
Duration: 4 Dec 20237 Dec 2023

Publication series

NameIEEE International Conference on Visual Communications and Image Processing
ISSN (Print)1018-8770
ISSN (electronic)2642-9357

Abstract

This paper explores the potential of a learned two-layer B-frame codec, known as TLZMC. TLZMC is one of the few early attempts that deviate from the hybrid-based coding architecture by skipping motion coding. With TLZMC, a low-resolution base layer is utilized to encode temporally unpredictable information. We address the question of whether adapting the base-layer bitrate can achieve better rate-distortion performance. We apply the feature map modulation technique to enable per-frame bitrate adaptation of the base layer. We then propose and compare three online search strategies for determining the base-layer rate parameter: per-level brute-force search, per-level greedy search, and per-frame greedy search. Experimental results show that our top-performing search strategy achieves 0.6%-15.8% Bjontegaard-Delta rate savings over TLZMC.

Keywords

    B-frame coding, content-Adaptive bit allocation, Learned video compression

ASJC Scopus subject areas

Cite this

Rate Adaptation for Learned Two-layer B-frame Coding without Signaling Motion Information. / Xie, Hong Sheng; Chen, Yi Hsin; Peng, Wen Hsiao et al.
2023 IEEE International Conference on Visual Communications and Image Processing, VCIP 2023. Institute of Electrical and Electronics Engineers Inc., 2023. (IEEE International Conference on Visual Communications and Image Processing).

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Xie, HS, Chen, YH, Peng, WH, Benjak, M & Ostermann, J 2023, Rate Adaptation for Learned Two-layer B-frame Coding without Signaling Motion Information. in 2023 IEEE International Conference on Visual Communications and Image Processing, VCIP 2023. IEEE International Conference on Visual Communications and Image Processing, Institute of Electrical and Electronics Engineers Inc., 2023 IEEE International Conference on Visual Communications and Image Processing, VCIP 2023, Jeju, Korea, Republic of, 4 Dec 2023. https://doi.org/10.1109/VCIP59821.2023.10402774
Xie, H. S., Chen, Y. H., Peng, W. H., Benjak, M., & Ostermann, J. (2023). Rate Adaptation for Learned Two-layer B-frame Coding without Signaling Motion Information. In 2023 IEEE International Conference on Visual Communications and Image Processing, VCIP 2023 (IEEE International Conference on Visual Communications and Image Processing). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/VCIP59821.2023.10402774
Xie HS, Chen YH, Peng WH, Benjak M, Ostermann J. Rate Adaptation for Learned Two-layer B-frame Coding without Signaling Motion Information. In 2023 IEEE International Conference on Visual Communications and Image Processing, VCIP 2023. Institute of Electrical and Electronics Engineers Inc. 2023. (IEEE International Conference on Visual Communications and Image Processing). doi: 10.1109/VCIP59821.2023.10402774
Xie, Hong Sheng ; Chen, Yi Hsin ; Peng, Wen Hsiao et al. / Rate Adaptation for Learned Two-layer B-frame Coding without Signaling Motion Information. 2023 IEEE International Conference on Visual Communications and Image Processing, VCIP 2023. Institute of Electrical and Electronics Engineers Inc., 2023. (IEEE International Conference on Visual Communications and Image Processing).
Download
@inproceedings{b60e35ecc2f3452d8c6d9a9dd3b42442,
title = "Rate Adaptation for Learned Two-layer B-frame Coding without Signaling Motion Information",
abstract = "This paper explores the potential of a learned two-layer B-frame codec, known as TLZMC. TLZMC is one of the few early attempts that deviate from the hybrid-based coding architecture by skipping motion coding. With TLZMC, a low-resolution base layer is utilized to encode temporally unpredictable information. We address the question of whether adapting the base-layer bitrate can achieve better rate-distortion performance. We apply the feature map modulation technique to enable per-frame bitrate adaptation of the base layer. We then propose and compare three online search strategies for determining the base-layer rate parameter: per-level brute-force search, per-level greedy search, and per-frame greedy search. Experimental results show that our top-performing search strategy achieves 0.6%-15.8% Bjontegaard-Delta rate savings over TLZMC.",
keywords = "B-frame coding, content-Adaptive bit allocation, Learned video compression",
author = "Xie, {Hong Sheng} and Chen, {Yi Hsin} and Peng, {Wen Hsiao} and Martin Benjak and Jorn Ostermann",
year = "2023",
doi = "10.1109/VCIP59821.2023.10402774",
language = "English",
isbn = "979-8-3503-5986-2",
series = "IEEE International Conference on Visual Communications and Image Processing",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "2023 IEEE International Conference on Visual Communications and Image Processing, VCIP 2023",
address = "United States",
note = "2023 IEEE International Conference on Visual Communications and Image Processing, VCIP 2023 ; Conference date: 04-12-2023 Through 07-12-2023",

}

Download

TY - GEN

T1 - Rate Adaptation for Learned Two-layer B-frame Coding without Signaling Motion Information

AU - Xie, Hong Sheng

AU - Chen, Yi Hsin

AU - Peng, Wen Hsiao

AU - Benjak, Martin

AU - Ostermann, Jorn

PY - 2023

Y1 - 2023

N2 - This paper explores the potential of a learned two-layer B-frame codec, known as TLZMC. TLZMC is one of the few early attempts that deviate from the hybrid-based coding architecture by skipping motion coding. With TLZMC, a low-resolution base layer is utilized to encode temporally unpredictable information. We address the question of whether adapting the base-layer bitrate can achieve better rate-distortion performance. We apply the feature map modulation technique to enable per-frame bitrate adaptation of the base layer. We then propose and compare three online search strategies for determining the base-layer rate parameter: per-level brute-force search, per-level greedy search, and per-frame greedy search. Experimental results show that our top-performing search strategy achieves 0.6%-15.8% Bjontegaard-Delta rate savings over TLZMC.

AB - This paper explores the potential of a learned two-layer B-frame codec, known as TLZMC. TLZMC is one of the few early attempts that deviate from the hybrid-based coding architecture by skipping motion coding. With TLZMC, a low-resolution base layer is utilized to encode temporally unpredictable information. We address the question of whether adapting the base-layer bitrate can achieve better rate-distortion performance. We apply the feature map modulation technique to enable per-frame bitrate adaptation of the base layer. We then propose and compare three online search strategies for determining the base-layer rate parameter: per-level brute-force search, per-level greedy search, and per-frame greedy search. Experimental results show that our top-performing search strategy achieves 0.6%-15.8% Bjontegaard-Delta rate savings over TLZMC.

KW - B-frame coding

KW - content-Adaptive bit allocation

KW - Learned video compression

UR - http://www.scopus.com/inward/record.url?scp=85184857498&partnerID=8YFLogxK

U2 - 10.1109/VCIP59821.2023.10402774

DO - 10.1109/VCIP59821.2023.10402774

M3 - Conference contribution

AN - SCOPUS:85184857498

SN - 979-8-3503-5986-2

T3 - IEEE International Conference on Visual Communications and Image Processing

BT - 2023 IEEE International Conference on Visual Communications and Image Processing, VCIP 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2023 IEEE International Conference on Visual Communications and Image Processing, VCIP 2023

Y2 - 4 December 2023 through 7 December 2023

ER -

By the same author(s)