Learning Image and Video Compression through Spatial-Temporal Energy Compaction

TL;DR

This paper introduces a novel learning-based image and video compression method that leverages spatial-temporal energy compaction, resulting in superior performance over traditional standards and existing learning methods, especially at high bit rates.

Contribution

It presents a convolutional autoencoder architecture for image compression and extends it to video by incorporating an interpolation loop and energy-based penalties, optimizing spatial-temporal energy distribution.

Findings

Outperforms latest image compression standards with MS-SSIM.

Significantly outperforms MPEG-4 in video compression.

Competitive with H.264, producing more visually pleasing results.

Abstract

Compression has been an important research topic for many decades, to produce a significant impact on data transmission and storage. Recent advances have shown a great potential of learning image and video compression. Inspired from related works, in this paper, we present an image compression architecture using a convolutional autoencoder, and then generalize image compression to video compression, by adding an interpolation loop into both encoder and decoder sides. Our basic idea is to realize spatial-temporal energy compaction in learning image and video compression. Thereby, we propose to add a spatial energy compaction-based penalty into loss function, to achieve higher image compression performance. Furthermore, based on temporal energy distribution, we propose to select the number of frames in one interpolation loop, adapting to the motion characteristics of video contents. Experimental results demonstrate that our proposed image compression outperforms the latest image compression standard with MS-SSIM quality metric, and provides higher performance compared with state-of-the-art learning compression methods at high bit rates, which benefits from our spatial energy compaction approach. Meanwhile, our proposed video compression approach with temporal energy compaction can significantly outperform MPEG-4 and is competitive with commonly used H.264. Both our image and video compression can produce more visually pleasant results than traditional standards.