Distributed Video Coding Based on Polar Codes

TL;DR

This paper introduces an improved distributed video coding scheme using polar codes, achieving better PSNR and faster decoding speeds compared to traditional LDPC-based methods.

Contribution

It proposes a novel polar coding approach for DVC, including efficient LLR modifications and nested polar code construction, enhancing video quality and decoding speed.

Findings

0.3 dB PSNR gain with modified LLRs over LDPCA codes

Up to 0.55 dB PSNR improvement for high-motion videos with polar codes

Decoding speed doubled with Tal-Vardy list decoder compared to belief propagation

Abstract

In this letter we present an improved distributed video coding (DVC) scheme based on polar coding techniques. Firstly, we adapt log-likelihood ratios (LLRs) for DVC with integer implementation of a discrete cosine transform (DCT). We propose a computationally efficient and numerically stable modification of these LLRs based on the simplified methods of polar codes decoding. We show that on average this approach provides 0.3 dB PSNR gain for DVC with LDPC accumulated (LDPCA) codes. Secondly, we introduce the nested shortened polar codes construction algorithm. We demonstrate that replacement of LDPCA by polar codes improves PSNR by 0.1 dB on average, whereas, for videos with relatively high motion level, the gain reaches up to 0.23, 0.39 and 0.55 dB for Group of Pictures (GOP) lengths 2, 4 and 8 frames, respectively. Finally, experimental results demonstrate that DVC with polar codes and Tal-Vardy list decoder operates up to two times faster than DVC with LDPCA code and belief propagation (BP) decoder.