Lifting Monocular Events to 3D Human Poses

被引:9
|
作者
Scarpellini, Gianluca [1 ,2 ]
Morerio, Pietro [1 ]
Del Bue, Alessio [1 ,3 ]
机构
[1] Ist Italiano Tecnol, Pattern Anal & Comp Vis, Genoa, Italy
[2] Univ Genoa, Genoa, Italy
[3] Ist Italiano Tecnol, Visual Geometry & Modelling, Genoa, Italy
关键词
D O I
10.1109/CVPRW53098.2021.00150
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper presents a novel 3D human pose estimation approach using a single stream of asynchronous events as input. Most of the state-of-the-art approaches solve this task with RGB cameras, however struggling when subjects are moving fast. On the other hand, event-based 3D pose estimation benefits from the advantages of event-cameras, especially their efficiency and robustness to appearance changes. Yet, finding human poses in asynchronous events is in general more challenging than standard RGB pose estimation, since little or no events are triggered in static scenes. Here we propose the first learning-based method for 3D human pose from a single stream of events. Our method consists of two steps. First, we process the event-camera stream to predict three orthogonal heatmaps per joint; each heatmap is the projection of of the joint onto one orthogonal plane. Next, we fuse the sets of heatmaps to estimate 3D localisation of the body joints. As a further contribution, we make available a new, challenging dataset for event-based human pose estimation by simulating events from the RGB Human3.6m dataset. Experiments demonstrate that our method achieves solid accuracy, narrowing the performance gap between standard RGB and event-based vision. The code is freely available at https://iit-pavis.github.io/lifting_events_to_3d_hpe.
引用
收藏
页码:1358 / 1368
页数:11
相关论文
共 50 条
  • [31] Monocular 3D Human Pose Estimation by Generation and Ordinal Ranking
    Sharma, Saurabh
    Varigonda, Pavan Teja
    Bindal, Prashast
    Sharma, Abhishek
    Jain, Arjun
    2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, : 2325 - 2334
  • [32] MoCapDeform: Monocular 3D Human Motion Capture in Deformable Scenes
    Li, Zhi
    Shimada, Soshi
    Schiele, Bernt
    Theobalt, Christian
    Golyanik, Vladislav
    2022 INTERNATIONAL CONFERENCE ON 3D VISION, 3DV, 2022, : 1 - 11
  • [33] Monocular 3D reconstruction of human motion in long action sequences
    Loy, G
    Eriksson, M
    Sullivan, J
    Carlsson, S
    COMPUTER VISION - ECCV 2004, PT 4, 2004, 2034 : 442 - 455
  • [34] Monocular 3D Human Pose Estimation by Predicting Depth on Joints
    Nie, Bruce Xiaohan
    Wei, Ping
    Zhu, Song-Chun
    2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2017, : 3467 - 3475
  • [35] Deep Kinematics Analysis for Monocular 3D Human Pose Estimation
    Xu, Jingwei
    Yu, Zhenbo
    Ni, Bingbing
    Yang, Jiancheng
    Yang, Xiaokang
    Zhang, Wenjun
    2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2020, : 896 - 905
  • [36] Chasing the Tail in Monocular 3D Human Reconstruction With Prototype Memory
    Rong, Yu
    Liu, Ziwei
    Loy, Chen Change
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2022, 31 : 2907 - 2919
  • [37] Double chain networks for monocular 3D human pose estimation
    Bai, Guihu
    Luo, Yanmin
    Pan, Xueliang
    Wang, Youjie
    Wang, Jia
    Guo, Jingming
    IMAGE AND VISION COMPUTING, 2022, 123
  • [38] Neural Monocular 3D Human Motion Capture with Physical Awareness
    Shimada, Soshi
    Golyanik, Vladislav
    Xu, Weipeng
    Perez, Patrick
    Theobalt, Christian
    ACM TRANSACTIONS ON GRAPHICS, 2021, 40 (04):
  • [39] 3D Human Motion Capture from Monocular Image Sequences
    Wandt, Bastian
    Ackermann, Hanno
    Rosenhahn, Bodo
    2015 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW), 2015,
  • [40] Recovering 3D Human Mesh From Monocular Images: A Survey
    Tian, Yating
    Zhang, Hongwen
    Liu, Yebin
    Wang, Limin
    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2023, 45 (12) : 15406 - 15425