MoSca: Dynamic Gaussian Fusion from Casual Videos via 4D Motion Scaffolds

1University of Pennsylvania, 2Stanford University, 3Archimedes Athena RC

MoSca reconstructs renderable 4D scenes from casual monocular videos in the wild.

Abstract

We introduce 4D Motion Scaffolds (MoSca), a neural information processing system designed to reconstruct and synthesize novel views of dynamic scenes from monocular videos captured casually in the wild. To address such a challenging and ill-posed inverse problem, we leverage prior knowledge from foundational vision models, lift the video data to a novel Motion Scaffold (MoSca) representation, which compactly and smoothly encodes the underlying motions / deformations. The scene geometry and appearance are then disentangled from the deformation field, and are encoded by globally fusing the Gaussians anchored onto the MoSca and optimized via Gaussian Splatting. Additionally, camera poses can be seamlessly initialized and refined during the dynamic rendering process, without the need for other pose estimation tools. Experiments demonstrate state-of-the-art performance on dynamic rendering benchmarks. The code will be released no later than the acceptance of this paper.

Video

Watch on Bilibili in China here.

Reconstructing DAVIS casual videos.

Reconstructing SORA generated videos.

Reconstructing movie clips.

Reconstructing internet robot videos.

BibTeX