构建73小时物理可信的人形机器人运动数据集，提升仿人行走稳定性。
PHUMA: Physically Reliable Humanoid Locomotion Dataset
- 两阶段流程：物理感知筛选+物理约束重定向，融合真人动捕与网络视频数据
- 在运动追踪任务中表现优于AMASS和Humanoid-X，零样本迁移至真实Unitree G1成功
- 解决浮空、穿模、足部滑移等物理伪影，适合强化学习与真实机器人部署
运动模仿是实现人形机器人仿人行走的有前景方法，但现有方法依赖高质量动捕数据集（如AMASS），这类数据稀缺且成本高，限制了可扩展性与多样性。近期研究尝试通过转化大规模网络视频来扩展数据，如Humanoid-X，但常出现浮空、穿透、足部滑移等物理伪影，影响稳定模仿。为此，我们提出PHUMA——一个通过两阶段流程（物理感知筛选与物理约束重定向）生成的物理可靠人形运动数据集，整合动捕与网络视频数据，形成73小时的高质量语料库。在运动追踪基准测试中，基于PHUMA训练的策略成功率高于AMASS与Humanoid-X，并成功实现零样本迁移至真实Unitree G1机器人。代码已公开于https://davian-robotics.github.io/PHUMA。
原文摘要 · Abstract (English)
Motion imitation is a promising approach for humanoid locomotion, enabling agents to acquire humanlike behaviors. Existing methods typically rely on high-quality motion capture datasets such as AMASS, but these are scarce and expensive, limiting scalability and diversity. Recent studies attempt to scale data collection by converting large-scale internet videos, exemplified by Humanoid-X. However, they often suffer from physical artifacts such as floating, penetration, and foot skating, which hinder stable imitation. To address this, we introduce PHUMA, a Physically Reliable HUMAnoid locomotion dataset produced by a two-stage pipeline combining physics-aware curation and physics-constrained retargeting, aggregating both motion capture and internet video into a physically reliable, 73-hour corpus. On motion tracking benchmarks, PHUMA-trained policies achieve higher success rates than those trained on AMASS and Humanoid-X, and successfully transfer zero-shot to a real Unitree G1. The code is available at https://davian-robotics.github.io/PHUMA.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。