arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.30293cs.CVcs.LG
收藏

EgoHand:基于第一视角视频的单目三维手部重建模型

EgoHand: Monocular 3D Hand Reconstruction from Egocentric Video

NVIDIA(英伟达)

Chao-Yeh Chen, Tiffany Chen, Lin Duan, Tianyi Hu, Yu-Hsiang Huang, Umar Iqbal, Jiefeng Li, Edy Lim, Yanchen Liu, Yihan Pang, Hang Qiu, Jun Saito, Shao Su, Qingsi Wang, Xiufeng Xie

AI总结:

本研究提出EgoHand单目三维手部重建模型,通过1400万张多源数据训练,在零样本评估中泛化能力强,优于相关基线,获用户偏好认可且支持免费商业使用。

摘要:

我们提出了EgoHand,一种在第一视角视频上达到了当前最优性能的单目三维手部重建模型。我们方法的核心是整理出了一个包含1400万张图像的高质量训练语料库,该语料库结合了合成数据、野外视频以及多摄像头动作捕捉数据,以平衡场景多样性与监督质量。该模型采用SOMA-Hand作为其参数化表示,使用Transformer编码器处理二维手部裁剪图像,同时融入相机线索和辅助二维分支,以提升几何一致性和图像对齐效果。在HOT3D、ARCTIC和H2O上的零样本评估显示出强大的域外泛化能力,在约9.69毫秒的推理延迟下,其关节位置、根平移和加速度误差均低于竞争基线。一项盲法用户偏好研究进一步表明,参与者更偏好EgoHand重建的手部姿态,而非Apple EgoDex和Ropedia Xperience-10M的重建结果。我们依据OpenMDW许可发布了EgoHand模型和源代码,允许对该模型进行免费商业使用。

英文摘要:

We introduce EgoHand, a monocular 3D hand reconstruction model that achieves state-of-the-art performance on egocentric video. Central to our approach is the curation of a high-quality training corpus of 14 million images, combining synthetic data, in-the-wild video, and multi-camera motion capture to balance scene diversity and supervision quality. The model adopts SOMA-Hand as its parametric representation and processes 2D hand crops with a transformer encoder, incorporating camera cues and an auxiliary 2D branch to improve geometric consistency and image alignment. Zero-shot evaluations on HOT3D, ARCTIC, and H2O demonstrate strong out-of-domain generalization, with lower joint-position, root-translation, and acceleration errors than competing baselines at an inference latency of approximately 9.01 ms. A blinded user preference study further shows that participants favor EgoHand's reconstructed hand poses over those from Apple EgoDex and Ropedia Xperience-10M. We release the EgoHand model and source code under the OpenMDW license, allowing free commercial use of the model.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑