Self-attentive 3D human pose and shape estimation from videos

被引：0

作者：

Chen, Yun-Chun ^{[1
]}

Piccirilli, Marco ^{[2
]}

Piramuthu, Robinson ^{[3
]}

Yang, Ming-Hsuan ^{[4
]}

机构：

[1] Department of Computer Science, University of Toronto, ON, Canada

[2] eBay Inc., San Jose, CA, United States

[3] Amazon Inc., Sunnyvale,CA, United States

[4] School of Engineering, University of California at Merced, CA, United States

来源：

Computer Vision and Image Understanding | 2021年 / 213卷

基金：

美国国家科学基金会;

关键词：

3D human pose estimation - Frame-based - Human pose - Human shapes - Long-range dependencies - Occlusion handling - Self-supervised learning - Shape estimation - Temporal information - Temporal prediction;

D O I：

103305

中图分类号：

学科分类号：

摘要：

引用