Unified Multivariate Gaussian Mixture for Efficient Neural Image Compression
Authors:
- 1 June 2022
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 17591-17600
Abstract
Modeling latent variables with priors and hyperpriors is an essential problem in variational image compression. Formally, trade-off between rate and distortion is handled well if priors and hyperpriors precisely describe latent variables. Current practices only adopt univariate priors and process each variable individually. However, we find inter-correlations and intra-correlations exist when observing latent variables in a vectorized perspective. These findings reveal visual redundancies to improve rate-distortion performance and parallel processing ability to speed up compression. This encourages us to propose a novel vectorized prior. Specifically, a multivariate Gaussian mixture is proposed with means and covariances to be estimated. Then, a novel probabilistic vector quantization is utilized to effectively approximate means, and remaining covariances are further induced to a unified mixture and solved by cascaded estimation without context models involved. Furthermore, code books involved in quantization are extended to multi-codebooks for complexity reduction, which formulates an efficient compression procedure. Extensive experiments on benchmark datasets against state-of-the-art indicate our model has better rate-distortion performance and an impressive 3.18x compression speed up, giving us the ability to perform real-time, high-quality variational image compression in practice. Our source code is publicly available at https://github.com/xiaosu-zhu/McQuic.Keywords
All Related Versions
Funding Information
- National Natural Science Foundation of China (62020106008,62122018,61772116,61872064)
This publication has 20 references indexed in Scilit:
- ImageNet classification with deep convolutional neural networksCommunications of the ACM, 2017
- Overview of the High Efficiency Video Coding (HEVC) StandardIEEE Transactions on Circuits and Systems for Video Technology, 2012
- Product Quantization for Nearest Neighbor SearchIEEE Transactions on Pattern Analysis and Machine Intelligence, 2010
- Rate‐Distortion Theory2003
- Image compression with neural networks – A surveySignal Processing: Image Communication, 1999
- The JPEG still picture compression standardCommunications of the ACM, 1991
- Image compression using address-vector quantizationIEEE Transactions on Communications, 1990
- Image Compression Based On Vector Quantization With Finite MemoryOptical Engineering, 1987
- Image Compression Using Adaptive Vector QuantizationIEEE Transactions on Communications, 1986
- The Jackknife Estimate of VarianceThe Annals of Statistics, 1981