Rate Distortion Characteristic Modeling for Neural Image Compression

TL;DR

This paper introduces a mathematical model for the rate-distortion behavior of neural image compression, enabling flexible bit-rate control with a single trained network, improving practical deployment.

Contribution

It proposes a novel model and plugin-in module to accurately describe and control the R-D characteristics of NIC across different bit-rates.

Findings

Achieves state-of-the-art continuous bit-rate coding performance

Enables single-network R-D control without multiple models

Facilitates practical deployment of NIC systems

Abstract

End-to-end optimized neural image compression (NIC) has obtained superior lossy compression performance recently. In this paper, we consider the problem of rate-distortion (R-D) characteristic analysis and modeling for NIC. We make efforts to formulate the essential mathematical functions to describe the R-D behavior of NIC using deep networks. Thus arbitrary bit-rate points could be elegantly realized by leveraging such model via a single trained network. We propose a plugin-in module to learn the relationship between the target bit-rate and the binary representation for the latent variable of auto-encoder. The proposed scheme resolves the problem of training distinct models to reach different points in the R-D space. Furthermore, we model the rate and distortion characteristic of NIC as a function of the coding parameter $\lambda$ respectively. Our experiments show our proposed method is easy to adopt and realizes state-of-the-art continuous bit-rate coding performance, which implies that our approach would benefit the practical deployment of NIC.