A Survey on Depth Ambiguity of 3D Human Pose Estimation

被引：7
|
作者：
Zhang, Siqi [1 ]
Wang, Chaofang [1 ]
Dong, Wenlong [1 ,2 ]
Fan, Bin [1 ,2 ]
机构：
[1] Tianjin Univ, Inst Disaster & Emergency Med, Tianjin 300072, Peoples R China
[2] Tianjin Univ, Wenzhou Safety Emergency Inst, Wenzhou 325000, Peoples R China
来源：
APPLIED SCIENCES-BASEL | 2022年 / 12卷 / 20期
关键词：
human pose estimation; depth ambiguity; deep learning; human key points positioning; human object detection;
D O I：
10.3390/app122010591
中图分类号：
O6 [化学];
学科分类号：
0703 ;
摘要：
Depth ambiguity is one of the main challenges of three-dimensional (3D) human pose estimation (HPE). The recent strategies of disambiguating have brought significant progress and remarkable breakthroughs in the field of 3D human pose estimation (3D HPE). This survey extensively reviews the causes and solutions of the depth ambiguity. The solutions are systematically classified into four categories: camera parameter constraints, temporal consistency constraints, kinematic constraints, and image cues constraints. This paper summarizes the performance comparison, challenges, main frameworks, and evaluation metrics, and discusses some promising future research directions.
引用
收藏
页数：10
相关论文
共 50 条
  • [1] Reducing Depth Ambiguity in 3D Human Pose and Body Shape Estimation
    Maruyama, Gakuto
    Kaneko, Naoshi
    Ito, Seiya
    Sumi, Kazuhiko
    [J]. FIFTEENTH INTERNATIONAL CONFERENCE ON QUALITY CONTROL BY ARTIFICIAL VISION, 2021, 11794
  • [2] 3D human pose estimation by depth map
    Wu, Jianzhai
    Hu, Dewen
    Xiang, Fengtao
    Yuan, Xingsheng
    Su, Jiongming
    [J]. VISUAL COMPUTER, 2020, 36 (07): : 1401 - 1410
  • [3] 3D human pose estimation by depth map
    Jianzhai Wu
    Dewen Hu
    Fengtao Xiang
    Xingsheng Yuan
    Jiongming Su
    [J]. The Visual Computer, 2020, 36 : 1401 - 1410
  • [4] A survey on monocular 3D human pose estimation
    Ji, Xiaopeng
    Fang, Qi
    Dong, Junting
    Shuai, Qing
    Jiang, Wen
    Zhou, Xiaowei
    [J]. Virtual Reality and Intelligent Hardware, 2020, 2 (06): : 471 - 500
  • [5] Ordinal Depth Supervision for 3D Human Pose Estimation
    Pavlakos, Georgios
    Zhou, Xiaowei
    Daniilidis, Kostas
    [J]. 2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, : 7307 - 7316
  • [6] DRPose3D: Depth Ranking in 3D Human Pose Estimation
    Wang, Min
    Chen, Xipeng
    Liu, Wentao
    Qian, Chen
    Lin, Liang
    Ma, Lizhuang
    [J]. PROCEEDINGS OF THE TWENTY-SEVENTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2018, : 978 - 984
  • [7] Monocular 3D Human Pose Estimation by Predicting Depth on Joints
    Nie, Bruce Xiaohan
    Wei, Ping
    Zhu, Song-Chun
    [J]. 2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2017, : 3467 - 3475
  • [8] Residual Pose: A Decoupled Approach for Depth-based 3D Human Pose Estimation
    Martinez-Gonzalez, Angel
    Villamizar, Michael
    Canevet, Olivier
    Odobez, Jean-Marc
    [J]. 2020 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS), 2020, : 10313 - 10318
  • [9] Towards Alleviating the Modeling Ambiguity of Unsupervised Monocular 3D Human Pose Estimation
    Yu, Zhenbo
    Ni, Bingbing
    Xu, Jingwei
    Wang, Junjie
    Zhao, Chenglong
    Zhang, Wenjun
    [J]. 2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021), 2021, : 8631 - 8640
  • [10] 3D Human Pose Estimation via Explicit Compositional Depth Maps
    Wu, Haiping
    Xiao, Bin
    [J]. THIRTY-FOURTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THE THIRTY-SECOND INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE AND THE TENTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2020, 34 : 12378 - 12385
12345