期刊
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017)
卷 -, 期 -, 页码 5759-5767出版社
IEEE
DOI: 10.1109/CVPR.2017.610
关键词
-
类别
资金
- NSF [1618903, 1208598]
- Intel Science and Technology Center for Visual Cloud Systems (ISTC-VCS)
- Amazon
- Direct For Computer & Info Scie & Enginr
- Div Of Information & Intelligent Systems [1618903] Funding Source: National Science Foundation
- Directorate For Engineering
- Div Of Civil, Mechanical, & Manufact Inn [1208598] Funding Source: National Science Foundation
We explore 3D human pose estimation from a single RGB image. While many approaches try to directly predict 3D pose from image measurements, we explore a simple architecture that reasons through intermediate 2D pose predictions. Our approach is based on two key observations (1) Deep neural nets have revolutionized 2D pose estimation, producing accurate 2D predictions even for poses with self-occlusions (2) Big-data sets of 3D mocap data are now readily available, making it tempting to lift predicted 2D poses to 3D through simple memorization (e.g., nearest neighbors). The resulting architecture is straightforward to implement with off-the-shelf 2D pose estimation systems and 3D mocap libraries. Importantly, we demonstrate that such methods outperform almost all state-of-the-art 3D pose estimation systems, most of which directly try to regress 3D pose from 2D measurements.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据