[论文] HumanTracker: Towards Comprehensive and Human-Aligned Motion Tracking ...

论文概要

研究领域: CV 作者: Dairu Liu, Zekun Qi, Jiayu Zeng, Ruixi Yu, Yu Guan, Yintianrun Zhang, Xuchuan Chen, Sikai Liang, Zekai Li, Chenghuai Lin, Xinqiang Yu, Wenyao Zhang, He Wang, Li Yi 发布时间: 2026-08-13 arXiv: 2608.13555

中文摘要

人形运动跟踪是遥操作和全身模仿的核心，但评估结果常与人们在视频中感知到的不一致。运动学误差平均了每帧姿势差异，但错过了最重要的物理伪影，特别是不稳定的支撑和不正确的接触，如脚滑和时机不当的触地。同时，广泛使用的测试套件规模较小，缺乏对富含接触的长时程行为进行压力测试所需的多样性。我们引入HumanTracker，使人形跟踪评估既与感知对齐又可扩展。HumanTracker基准包含来自多位专业表演者约153小时的光学运动轨迹，组织为四个运动家族，带有文本标签用于细粒度诊断。我们进一步提出HumanScore，一种在包含24K个运动的12K个运动对上训练的偏好对齐指标。在代表性的最先进跟踪器中，HumanScore更好地预测人类偏好，并揭示运动学指标经常遗漏的接触和稳定性失败。

原文摘要

Humanoid motion tracking is central to teleoperation and whole-body imitation, yet evaluation often disagrees with what people perceive in videos. Kinematic errors average per-frame pose differences but miss the physical artifacts that matter most, particularly unstable support and incorrect contacts such as foot skating and mistimed touch-downs. Meanwhile, widely used test suites are small and lack the diversity needed to stress contact-rich, long-horizon behaviors. We introduce HumanTracker to make humanoid tracking evaluation both perceptually aligned and scalable. The HumanTracker benchmark contains approximately 153 hours of optical motion trajectories from multiple professional performers, organized into four motion families with text labels for fine-grained diagnosis. We further propose HumanScore, a preference-aligned metric trained on 12K motion pairs containing 24K motions. Across representative state-of-the-art trackers, HumanScore better predicts human preferences and reveals contact and stability failures that kinematic metrics often miss.

--- *自动采集于 2026-08-15*

#论文 #arXiv #CV #小凯