Learned Image Compression with Soft Bit-Based Rate-Distortion Optimization

David Alexandre, Chih Peng Chang, Wen Hsiao Peng, Hsueh Ming Hang

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

Abstract

This paper introduces the notion of soft bits to address the rate-distortion optimization for learning-based image compression. Recent methods for such compression train an autoencoder end-to-end with an objective to strike a balance between distortion and rate. They are faced with the zero gradient issue due to quantization and the difficulty of estimating the rate accurately. Inspired by soft quantization, we represent quantization indices of feature maps with differentiable soft bits. This allows us to couple tightly the rate estimation with context-adaptive binary arithmetic coding. It also provides a differentiable distortion objective function. Experimental results show that our approach achieves the state-of-the-art compression performance among the learning-based schemes in terms of MS-SSIM and PSNR.

Original languageEnglish
Title of host publication2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings
PublisherIEEE Computer Society
Pages1715-1719
Number of pages5
ISBN (Electronic)9781538662496
DOIs
StatePublished - Sep 2019
Event26th IEEE International Conference on Image Processing, ICIP 2019 - Taipei, Taiwan
Duration: 22 Sep 201925 Sep 2019

Publication series

NameProceedings - International Conference on Image Processing, ICIP
Volume2019-September
ISSN (Print)1522-4880

Conference

Conference26th IEEE International Conference on Image Processing, ICIP 2019
CountryTaiwan
CityTaipei
Period22/09/1925/09/19

Keywords

  • Autoencoder
  • Deep Learning
  • Image Compression
  • Soft Bits

Fingerprint Dive into the research topics of 'Learned Image Compression with Soft Bit-Based Rate-Distortion Optimization'. Together they form a unique fingerprint.

Cite this