Robust Dual Gaussian Splatting for Immersive Human-centric Volumetric Videos

TL;DR

This paper introduces DualGS, a Gaussian-based method for real-time, high-fidelity volumetric human performance playback that significantly reduces data size and enhances temporal coherence for immersive VR experiences.

Contribution

The paper presents a novel dual Gaussian representation that separately models motion and appearance, enabling efficient compression and high-quality rendering of volumetric videos.

Findings

Achieves up to 120x compression ratio.

Requires only about 350KB per frame.

Enables photo-realistic VR experiences of human performances.

Abstract

Volumetric video represents a transformative advancement in visual media, enabling users to freely navigate immersive virtual experiences and narrowing the gap between digital and real worlds. However, the need for extensive manual intervention to stabilize mesh sequences and the generation of excessively large assets in existing workflows impedes broader adoption. In this paper, we present a novel Gaussian-based approach, dubbed \textit{DualGS}, for real-time and high-fidelity playback of complex human performance with excellent compression ratios. Our key idea in DualGS is to separately represent motion and appearance using the corresponding skin and joint Gaussians. Such an explicit disentanglement can significantly reduce motion redundancy and enhance temporal coherence. We begin by initializing the DualGS and anchoring skin Gaussians to joint Gaussians at the first frame. Subsequently, we employ a coarse-to-fine training strategy for frame-by-frame human performance modeling. It includes a coarse alignment phase for overall motion prediction as well as a fine-grained optimization for robust tracking and high-fidelity rendering. To integrate volumetric video seamlessly into VR environments, we efficiently compress motion using entropy encoding and appearance using codec compression coupled with a persistent codebook. Our approach achieves a compression ratio of up to 120 times, only requiring approximately 350KB of storage per frame. We demonstrate the efficacy of our representation through photo-realistic, free-view experiences on VR headsets, enabling users to immersively watch musicians in performance and feel the rhythm of the notes at the performers' fingertips.