Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

TRACE: Learning 3D Gaussian Physical Dynamics from Multi-view Videos

About

In this paper, we aim to model 3D scene geometry, appearance, and physical information just from dynamic multi-view videos in the absence of any human labels. By leveraging physics-informed losses as soft constraints or integrating simple physics models into neural nets, existing works often fail to learn complex motion physics, or doing so requires additional labels such as object types or masks. We propose a new framework named TRACE to model the motion physics of complex dynamic 3D scenes. The key novelty of our method is that, by formulating each 3D point as a rigid particle with size and orientation in space, we directly learn a translation rotation dynamics system for each particle, explicitly estimating a complete set of physical parameters to govern the particle's motion over time. Extensive experiments on three existing dynamic datasets and one newly created challenging synthetic datasets demonstrate the extraordinary performance of our method over baselines in the task of future frame extrapolation. A nice property of our framework is that multiple objects or parts can be easily segmented just by clustering the learned physical parameters.

Jinxi Li, Ziyang Song, Bo Yang• 2025

Related benchmarks

TaskDatasetResultRank
Future frame extrapolationDynamic Indoor Scene Dataset
PSNR34.824
24
Future frame extrapolationDynamic Object Dataset
PSNR31.597
22
Future frame extrapolationNVIDIA Dynamic Scene Skating
PSNR29.409
12
Future frame extrapolationNVIDIA Dynamic Scene Truck
PSNR29.252
12
Novel view interpolationNVIDIA Dynamic Scene Truck
PSNR28.316
12
Novel view interpolationNVIDIA Dynamic Scene Skating
PSNR28.198
12
Future frame extrapolationDynamic Multipart (test)
PSNR33.481
9
3D Scene ReconstructionDynamic Indoor Scene Dining Table
PSNR35.58
4
3D Scene ReconstructionDynamic Indoor Scene Chessboard
PSNR34.63
4
3D Scene ReconstructionDynamic Indoor Scene Darkroom
PSNR37.774
4
Showing 10 of 12 rows

Other info

Follow for update