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

2026-02-17

AI Summary

Existing methods for humanoid motion synthesis rely on expensive motion capture data and struggle to effectively couple with environmental geometry, often resulting in physically inconsistent behaviors such as foot sliding or mesh penetration on complex terrains. This work proposes the first framework that jointly reconstructs human motion and 3D scene geometry from monocular video, enabling geometry-aware dynamic locomotion through a motion–terrain coupling learning mechanism. By integrating state-of-the-art 3D vision models, kinematic consistency optimization, and contact-invariant retargeting, the approach extracts high-fidelity motion features from noisy visual inputs without requiring motion capture data. The method demonstrates robust, highly dynamic humanoid locomotion across diverse challenging terrains, establishing the feasibility of training sophisticated physical interaction capabilities using only consumer-grade monocular sensors.

Technology Category

• Computer Vision: Motion & Tracking
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

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.

Problem

• humanoid motion control
• motion-terrain interaction
• 3D scene reconstruction
• physical consistency
• motion capture data

Innovation

• 3D scene reconstruction
• humanoid motion learning
• contact-invariant retargeting
• kinematic consistency optimization
• embodied intelligence