Explicit Occlusion Reasoning for Multi-person 3D Human Pose Estimation

被引:10
|
作者
Liu, Qihao [1 ]
Zhang, Yi [1 ]
Bai, Song [2 ]
Yuille, Alan [1 ]
机构
[1] Johns Hopkins Univ, Baltimore, MD USA
[2] ByteDance, Singapore, Singapore
来源
关键词
Human pose estimation; 3D from a single image;
D O I
10.1007/978-3-031-20065-6_29
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Occlusion poses a great threat to monocular multi-person 3D human pose estimation due to large variability in terms of the shape, appearance, and position of occluders. While existing methods try to handle occlusion with pose priors/constraints, data augmentation, or implicit reasoning, they still fail to generalize to unseen poses or occlusion cases and may make large mistakes when multiple people are present. Inspired by the remarkable ability of humans to infer occluded joints from visible cues, we develop a method to explicitly model this process that significantly improves bottom-up multi-person human pose estimation with or without occlusions. First, we split the task into two subtasks: visible keypoints detection and occluded keypoints reasoning, and propose a Deeply Supervised Encoder Distillation (DSED) network to solve the second one. To train our model, we propose a Skeleton-guided human Shape Fitting (SSF) approach to generate pseudo occlusion labels on the existing datasets, enabling explicit occlusion reasoning. Experiments show that explicitly learning from occlusions improves human pose estimation. In addition, exploiting feature-level information of visible joints allows us to reason about occluded joints more accurately. Our method outperforms both the state-of-the-art top-down and bottom-up methods on several benchmarks.
引用
收藏
页码:497 / 517
页数:21
相关论文
共 50 条
  • [41] Multi-person 3D pose estimation from a single image captured by a fisheye camera
    Zhang, Yahui
    You, Shaodi
    Karaoglu, Sezer
    Gevers, Theo
    [J]. COMPUTER VISION AND IMAGE UNDERSTANDING, 2022, 222
  • [42] Fast and Robust Multi-Person 3D Pose Estimation and Tracking From Multiple Views
    Dong, Junting
    Fang, Qi
    Jiang, Wen
    Yang, Yurou
    Huang, Qixing
    Bao, Hujun
    Zhou, Xiaowei
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2022, 44 (10) : 6981 - 6992
  • [43] Multi-person 3D pose estimation from 3D cloud data using 3D convolutional neural networks
    Vasileiadis, Manolis
    Bouganis, Christos-Savvas
    Tzovaras, Dimitrios
    [J]. COMPUTER VISION AND IMAGE UNDERSTANDING, 2019, 185 : 12 - 23
  • [44] Pose Knowledge Transfer for multi-person pose estimation
    Li, Buwei
    Ji, Yi
    Li, Ying
    Xu, Yunlong
    Liu, Chunping
    [J]. SIGNAL IMAGE AND VIDEO PROCESSING, 2022, 16 (02) : 321 - 328
  • [45] Pose Partition Networks for Multi-person Pose Estimation
    Nie, Xuecheng
    Feng, Jiashi
    Xing, Junliang
    Yan, Shuicheng
    [J]. COMPUTER VISION - ECCV 2018, PT V, 2018, 11209 : 705 - 720
  • [46] Pose Knowledge Transfer for multi-person pose estimation
    Buwei Li
    Yi Ji
    Ying Li
    Yunlong Xu
    Chunping Liu
    [J]. Signal, Image and Video Processing, 2022, 16 : 321 - 328
  • [47] Multi-Person Pose Estimation With Human Detection: A Parallel Approach
    Van-Thanh Hoang
    Jo, Kang-Hyun
    [J]. IECON 2018 - 44TH ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY, 2018, : 3269 - 3272
  • [48] Depth-Aware Multi-Person 3D Pose Estimation With Multi-Scale Waterfall Representations
    Shen, Tianyu
    Li, Deqi
    Wang, Fei-Yue
    Huang, Hua
    [J]. IEEE TRANSACTIONS ON MULTIMEDIA, 2023, 25 : 1439 - 1451
  • [49] Multi-person Human Pose Estimation Based on Deformable Convolution
    Zhao, Yunxiao
    Qian, Yuhua
    Wang, Keqi
    [J]. Moshi Shibie yu Rengong Zhineng/Pattern Recognition and Artificial Intelligence, 2020, 33 (10): : 944 - 950
  • [50] Graph-Based 3D Multi-Person Pose Estimation Using Multi-View Images
    Wu, Size
    Jin, Sheng
    Liu, Wentao
    Bai, Lei
    Qian, Chen
    Liu, Dong
    Ouyang, Wanli
    [J]. 2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021), 2021, : 11128 - 11137