Weakly-Supervised 3D Human Pose Learning via Multi-view Images in the Wild

被引:83
作者
Iqbal, Umar [1 ]
Molchanov, Pavlo [1 ]
Kautz, Jan [1 ]
机构
[1] NVIDIA, Santa Clara, CA 95051 USA
来源
2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) | 2020年
关键词
D O I
10.1109/CVPR42600.2020.00529
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
One major challenge for monocular 3D human pose estimation in-the-wild is the acquisition of training data that contains unconstrained images annotated with accurate 3D poses. In this paper, we address this challenge by proposing a weakly-supervised approach that does not require 3D annotations and learns to estimate 3D poses from unlabeled multi-view data, which can be acquired easily in in-the-wild environments. We propose a novel end-to-end learning framework that enables weakly-supervised training using multi-view consistency. Since multi-view consistency is prone to degenerated solutions, we adopt a 2.5D pose representation and propose a novel objective function that can only be minimized when the predictions of the trained model are consistent and plausible across all camera views. We evaluate our proposed approach on two large scale datasets (Human3.6M and MPII-INF-3DHP) where it achieves state-of-the-art performance among semi-/weakly-supervised methods.
引用
收藏
页码:5242 / 5251
页数:10
相关论文
共 52 条
[1]  
Martinez AA, 2017, INT SYMP COMPUT EDUC
[2]   2D Human Pose Estimation: New Benchmark and State of the Art Analysis [J].
Andriluka, Mykhaylo ;
Pishchulin, Leonid ;
Gehler, Peter ;
Schiele, Bernt .
2014 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2014, :3686-3693
[3]  
[Anonymous], 2016, 3DV
[4]  
[Anonymous], 2019, CVPR
[5]  
[Anonymous], 2018, ECCV
[6]  
[Anonymous], 2014, CVPR
[7]  
[Anonymous], 2016, ECCV
[8]  
[Anonymous], 2014, ACCV
[9]  
[Anonymous], 2016, BMVC
[10]  
[Anonymous], 2017, CVPR