Neural network-based arithmetic coding of intra prediction modes in HEVC

TL;DR

This paper introduces a neural network-based arithmetic coding method for intra prediction modes in HEVC, replacing traditional binarization and handcrafted context models, achieving up to 9.9% bits savings.

Contribution

It proposes a neural network-driven probability estimation approach for intra mode coding in HEVC, improving compression efficiency over conventional methods.

Findings

Achieves up to 9.9% bits saving compared to CABAC.

Uses CNN to directly estimate probability distributions.

Eliminates the need for binarization and handcrafted context models.

Abstract

In both H.264 and HEVC, context-adaptive binary arithmetic coding (CABAC) is adopted as the entropy coding method. CABAC relies on manually designed binarization processes as well as handcrafted context models, which may restrict the compression efficiency. In this paper, we propose an arithmetic coding strategy by training neural networks, and make preliminary studies on coding of the intra prediction modes in HEVC. Instead of binarization, we propose to directly estimate the probability distribution of the 35 intra prediction modes with the adoption of a multi-level arithmetic codec. Instead of handcrafted context models, we utilize convolutional neural network (CNN) to perform the probability estimation. Simulation results show that our proposed arithmetic coding leads to as high as 9.9% bits saving compared with CABAC.