arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2004.03143cs.CV
收藏

预测摄像机视点提高了3D人体姿态估计的跨数据集泛化能力

Predicting Camera Viewpoint Improves Cross-dataset Generalization for 3D Human Pose Estimation

University of California, Irvine(加州大学欧文分校)

Zhe Wang, Daeyun Shin, Charless C. Fowlkes

摘要:

随着大规模地面真实运动捕获数据集的出现,三维人体姿态的单目估计引起了越来越多的关注。然而,可用的训练数据的多样性是有限的,目前还不清楚在何种程度上的方法概括外的特定datasets,他们是trainedon.In这项工作中,我们进行了系统的研究的多样性和偏见存在于特定的数据集和跨数据集概括其效果在一个汇编的5姿势datasets。我们特别专注于系统的差异,相对于身体为中心的坐标系的相机视点的分布。基于这一观察,我们提出了一个辅助任务,预测相机的观点,除了姿势。我们发现,经过训练以联合预测视点和姿势的模型系统地显示出显着改善的跨数据集泛化。

英文摘要:

Monocular estimation of 3d human pose has attracted increased attention with the availability of large ground-truth motion capture datasets. However, the diversity of training data available is limited and it is not clear to what extent methods generalize outside the specific datasets they are trained on. In this work we carry out a systematic study of the diversity and biases present in specific datasets and its effect on cross-dataset generalization across a compendium of 5 pose datasets. We specifically focus on systematic differences in the distribution of camera viewpoints relative to a body-centered coordinate frame. Based on this observation, we propose an auxiliary task of predicting the camera viewpoint in addition to pose. We find that models trained to jointly predict viewpoint and pose systematically show significantly improved cross-dataset generalization.

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。

↑