HEVC/H.265 coding unit split decision using deep reinforcement learning

Chia Hua Chung, Wen Hsiao Peng, Jun Hao Hu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Scopus citations

Abstract

The video coding community has long been seeking more effective rate-distortion optimization techniques than the widely adopted greedy approach. The difficulty arises when we need to predict how the coding mode decision made in one stage would affect subsequent decisions and thus the overall coding performance. Taking a data-driven approach, we introduce in this paper deep reinforcement learning (RL) as a mechanism for the coding unit (CU) split decision in HEVC/H.265. We propose to regard the luminance samples of a CU together with the quantization parameter as its state, the split decision as an action, and the reduction in ratedistortion cost relative to keeping the current CU intact as the immediate reward. Based on the Q-learning algorithm, we learn a convolutional neural network to approximate the ratedistortion cost reduction of each possible state-action pair. The proposed scheme performs compatibly with the current full rate-distortion optimization scheme in HM-16.15, incurring a 2.5% average BD-rate loss. While also performing similarly to a conventional scheme that treats the split decision as a binary classification problem, our scheme can additionally quantify the rate-distortion cost reduction, enabling more applications.

Original languageEnglish
Title of host publication2017 International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2017 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages570-575
Number of pages6
ISBN (Electronic)9781538621592
DOIs
StatePublished - 2 Jul 2017
Event25th International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2017 - Xiamen, China
Duration: 6 Nov 20179 Nov 2017

Publication series

Name2017 International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2017 - Proceedings
Volume2018-January

Conference

Conference25th International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2017
CountryChina
CityXiamen
Period6/11/179/11/17

Keywords

  • deep reinforcement learning
  • HEVC/H.265
  • mode decision

Fingerprint Dive into the research topics of 'HEVC/H.265 coding unit split decision using deep reinforcement learning'. Together they form a unique fingerprint.

  • Cite this

    Chung, C. H., Peng, W. H., & Hu, J. H. (2017). HEVC/H.265 coding unit split decision using deep reinforcement learning. In 2017 International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2017 - Proceedings (pp. 570-575). (2017 International Symposium on Intelligent Signal Processing and Communication Systems, ISPACS 2017 - Proceedings; Vol. 2018-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISPACS.2017.8266543