Learning-Based Video Coding with Joint Deep Compression and Enhancement

TL;DR

This paper introduces an end-to-end deep video codec with joint compression and enhancement modules, utilizing a dual-path GAN for detail reconstruction, achieving significant bitrate reduction and outperforming existing codecs.

Contribution

The paper presents a novel joint deep video coding framework with a dual-path GAN for detail reconstruction and integrated modules for compression and enhancement, improving rate-distortion performance.

Findings

Reduces bit-per-pixel by up to 54.92% at same PSNR/MS-SSIM

Outperforms state-of-the-art deep video codecs

Enhances rate-distortion efficiency through joint training

Abstract

The end-to-end learning-based video compression has attracted substantial attentions by paving another way to compress video signals as stacked visual features. This paper proposes an efficient end-to-end deep video codec with jointly optimized compression and enhancement modules (JCEVC). First, we propose a dual-path generative adversarial network (DPEG) to reconstruct video details after compression. An $\alpha$-path facilitates the structure information reconstruction with a large receptive field and multi-frame references, while a $\beta$-path facilitates the reconstruction of local textures. Both paths are fused and co-trained within a generative-adversarial process. Second, we reuse the DPEG network in both motion compensation and quality enhancement modules, which are further combined with other necessary modules to formulate our JCEVC framework. Third, we employ a joint training of deep video compression and enhancement that further improves the rate-distortion (RD) performance of compression. Compared with x265 LDP very fast mode, our JCEVC reduces the average bit-per-pixel (bpp) by 39.39\%/54.92\% at the same PSNR/MS-SSIM, which outperforms the state-of-the-art deep video codecs by a considerable margin.