Generative Human Video Compression with Multi-granularity Temporal Trajectory Factorization

TL;DR

This paper introduces a multi-granularity temporal trajectory factorization framework for efficient generative human video compression, enabling high-quality reconstruction with low bandwidth by compact motion representation and resolution adaptability.

Contribution

It presents a novel motion factorization strategy and a resolution-expandable generative module, improving compression efficiency and reconstruction robustness over existing methods.

Findings

Outperforms state-of-the-art video coding standards in quality.

Achieves higher compression efficiency for human videos.

Demonstrates robustness and flexibility in resolution adaptation.

Abstract

In this paper, we propose a novel Multi-granularity Temporal Trajectory Factorization framework for generative human video compression, which holds great potential for bandwidth-constrained human-centric video communication. In particular, the proposed motion factorization strategy can facilitate to implicitly characterize the high-dimensional visual signal into compact motion vectors for representation compactness and further transform these vectors into a fine-grained field for motion expressibility. As such, the coded bit-stream can be entailed with enough visual motion information at the lowest representation cost. Meanwhile, a resolution-expandable generative module is developed with enhanced background stability, such that the proposed framework can be optimized towards higher reconstruction robustness and more flexible resolution adaptation. Experimental results show that proposed method outperforms latest generative models and the state-of-the-art video coding standard Versatile Video Coding (VVC) on both talking-face videos and moving-body videos in terms of both objective and subjective quality. The project page can be found at https://github.com/xyzysz/Extreme-Human-Video-Compression-with-MTTF.