Coherent Human-Scene Reconstruction from Multi-Person Multi-View Video in a Single Pass

TL;DR

CHROMM is a unified neural framework for multi-person, multi-view 3D human and scene reconstruction that operates in a single pass, integrating geometric and human priors for fast, accurate results.

Contribution

The paper introduces CHROMM, a novel end-to-end neural network that jointly estimates cameras, scene point clouds, and human meshes from multi-view videos without external modules.

Findings

Achieves competitive accuracy in human motion and pose estimation.

Runs over 8 times faster than previous optimization-based methods.

Effectively handles scale discrepancies and multi-person association.

Abstract

Recent advances in 3D foundation models have led to growing interest in reconstructing humans and their surrounding environments. However, most existing approaches focus on monocular inputs, and extending them to multi-view settings requires additional overhead modules or preprocessed data. To this end, we present CHROMM, a unified framework that jointly estimates cameras, scene point clouds, and human meshes from multi-person multi-view videos without relying on external modules or preprocessing. We integrate strong geometric and human priors from Pi3X and Multi-HMR into a single trainable neural network architecture, and introduce a scale adjustment module to solve the scale discrepancy between humans and the scene. We also introduce a multi-view fusion strategy to aggregate per-view estimates into a single representation at test-time. Finally, we propose a geometry-based multi-person association method, which is more robust than appearance-based approaches. Experiments on EMDB, RICH, EgoHumans, and EgoExo4D show that CHROMM achieves competitive performance in global human motion and multi-view pose estimation while running over 8x faster than prior optimization-based multi-view approaches. Project page: https://nstar1125.github.io/chromm.