Conditional Probability Models for Deep Image Compression

TL;DR

This paper introduces a novel approach for deep image compression by modeling the entropy of latent representations with a context-based 3D-CNN, improving rate-distortion performance in auto-encoders.

Contribution

It proposes a new technique to directly model the entropy of latent representations using a context model, enhancing deep image compression methods.

Findings

Achieves state-of-the-art MS-SSIM performance in image compression.

Uses a 3D-CNN context model to learn dependencies in latent space.

Improves rate-distortion trade-off in auto-encoder based compression.

Abstract

Deep Neural Networks trained as image auto-encoders have recently emerged as a promising direction for advancing the state-of-the-art in image compression. The key challenge in learning such networks is twofold: To deal with quantization, and to control the trade-off between reconstruction error (distortion) and entropy (rate) of the latent image representation. In this paper, we focus on the latter challenge and propose a new technique to navigate the rate-distortion trade-off for an image compression auto-encoder. The main idea is to directly model the entropy of the latent representation by using a context model: A 3D-CNN which learns a conditional probability model of the latent distribution of the auto-encoder. During training, the auto-encoder makes use of the context model to estimate the entropy of its representation, and the context model is concurrently updated to learn the dependencies between the symbols in the latent representation. Our experiments show that this approach, when measured in MS-SSIM, yields a state-of-the-art image compression system based on a simple convolutional auto-encoder.