A Hybrid Deep Animation Codec for Low-bitrate Video Conferencing

TL;DR

This paper introduces a layered hybrid deep animation codec for low-bitrate video conferencing that combines facial animation with a low bitrate conventional video stream, significantly improving compression efficiency.

Contribution

It proposes a novel layered hybrid coding scheme that integrates facial animation with traditional video compression to enhance performance at higher bitrates.

Findings

Achieves over 30% BD-Rate gains on video conferencing sequences.

Extends the effective bitrate range of facial animation codecs.

Demonstrates consistent performance improvements across a large dataset.

Abstract

Deep generative models, and particularly facial animation schemes, can be used in video conferencing applications to efficiently compress a video through a sparse set of keypoints, without the need to transmit dense motion vectors. While these schemes bring significant coding gains over conventional video codecs at low bitrates, their performance saturates quickly when the available bandwidth increases. In this paper, we propose a layered, hybrid coding scheme to overcome this limitation. Specifically, we extend a codec based on facial animation by adding an auxiliary stream consisting of a very low bitrate version of the video, obtained through a conventional video codec (e.g., HEVC). The animated and auxiliary videos are combined through a novel fusion module. Our results show consistent average BD-Rate gains in excess of -30% on a large dataset of video conferencing sequences, extending the operational range of bitrates of a facial animation codec alone