HSImul3R: Physics-in-the-Loop Reconstruction of Simulation-Ready Human-Scene Interactions

TL;DR

HSImul3R is a novel framework that achieves physically stable, simulation-ready 3D reconstructions of human-scene interactions from casual captures, bridging perception and physics for robotics applications.

Contribution

It introduces a physics-grounded bi-directional optimization pipeline and a new benchmark for stable, simulation-ready human-scene interaction reconstructions.

Findings

Produces the first stable, simulation-ready HSI reconstructions.

Enables direct deployment to real-world humanoid robots.

Bridges perception-simulation gap in human-scene modeling.

Abstract

We present HSImul3R, a unified framework for simulation-ready 3D reconstruction of human-scene interactions (HSI) from casual captures, including sparse-view images and monocular videos. Existing methods suffer from a perception-simulation gap: visually plausible reconstructions often violate physical constraints, leading to instability in physics engines and failure in embodied AI applications. To bridge this gap, we introduce a physically-grounded bi-directional optimization pipeline that treats the physics simulator as an active supervisor to jointly refine human dynamics and scene geometry. In the forward direction, we employ Scene-targeted Reinforcement Learning to optimize human motion under dual supervision of motion fidelity and contact stability. In the reverse direction, we propose Direct Simulation Reward Optimization, which leverages simulation feedback on gravitational stability and interaction success to refine scene geometry. We further present HSIBench, a new benchmark with diverse objects and interaction scenarios. Extensive experiments demonstrate that HSImul3R produces the first stable, simulation-ready HSI reconstructions and can be directly deployed to real-world humanoid robots.