One-shot Humanoid Whole-body Motion Learning
This addresses the labor-intensive and costly data collection problem in robotics for enabling human-like behaviors, representing an incremental improvement over existing methods.
The paper tackles the challenge of learning whole-body humanoid motion with minimal data by proposing a method that trains motion policies using only one non-walking sample plus walking motions, achieving superior performance on the CMU MoCap dataset.
Whole-body humanoid motion represents a cornerstone challenge in robotics, integrating balance, coordination, and adaptability to enable human-like behaviors. However, existing methods typically require multiple training samples per motion category, rendering the collection of high-quality human motion datasets both labor-intensive and costly. To address this, we propose a novel approach that trains effective humanoid motion policies using only a single non-walking target motion sample alongside readily available walking motions. The core idea lies in leveraging order-preserving optimal transport to compute distances between walking and non-walking sequences, followed by interpolation along geodesics to generate new intermediate pose skeletons, which are then optimized for collision-free configurations and retargeted to the humanoid before integration into a simulated environment for policy training via reinforcement learning. Experimental evaluations on the CMU MoCap dataset demonstrate that our method consistently outperforms baselines, achieving superior performance across metrics. Code will be released upon acceptance.