MeshMimic: Geometry-Aware Humanoid Motion Learning through 3D Scene Reconstruction

TL;DR

MeshMimic is a novel framework that combines 3D scene reconstruction and motion learning from video to enable humanoid robots to perform complex, terrain-aware behaviors without expensive motion capture data.

Contribution

It introduces a geometry-aware motion learning approach that integrates 3D scene understanding with humanoid control, reducing reliance on costly datasets and improving physical interaction fidelity.

Findings

Achieves robust humanoid motion on diverse terrains

Utilizes low-cost monocular sensors for training data

Demonstrates improved physical interaction accuracy

Abstract

Humanoid motion control has witnessed significant breakthroughs in recent years, with deep reinforcement learning (RL) emerging as a primary catalyst for achieving complex, human-like behaviors. However, the high dimensionality and intricate dynamics of humanoid robots make manual motion design impractical, leading to a heavy reliance on expensive motion capture (MoCap) data. These datasets are not only costly to acquire but also frequently lack the necessary geometric context of the surrounding physical environment. Consequently, existing motion synthesis frameworks often suffer from a decoupling of motion and scene, resulting in physical inconsistencies such as contact slippage or mesh penetration during terrain-aware tasks. In this work, we present MeshMimic, an innovative framework that bridges 3D scene reconstruction and embodied intelligence to enable humanoid robots to learn coupled "motion-terrain" interactions directly from video. By leveraging state-of-the-art 3D vision models, our framework precisely segments and reconstructs both human trajectories and the underlying 3D geometry of terrains and objects. We introduce an optimization algorithm based on kinematic consistency to extract high-quality motion data from noisy visual reconstructions, alongside a contact-invariant retargeting method that transfers human-environment interaction features to the humanoid agent. Experimental results demonstrate that MeshMimic achieves robust, highly dynamic performance across diverse and challenging terrains. Our approach proves that a low-cost pipeline utilizing only consumer-grade monocular sensors can facilitate the training of complex physical interactions, offering a scalable path toward the autonomous evolution of humanoid robots in unstructured environments.