Unsupervised 3D Human Pose Estimation in Multi-view-multi-pose Video

被引:0
|
作者
Sun, Cheng [1 ]
Thomas, Diego [1 ]
Kawasaki, Hiroshi [1 ]
机构
[1] Kyushu Univ, Fukuoka, Japan
来源
2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR) | 2021年
关键词
D O I
10.1109/ICPR48806.2021.9412270
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
3D human pose estimation from a single 2D video is an extremely difficult task because computing 3D geometry from 2D images is an ill-posed problem. Recent popular solutions adopt fully-supervised learning strategy, which requires to train a deep network on a large-scale ground truth dataset of 3D poses and 2D images. However, such a large-scale dataset with natural images does not exist, which limits the usability of existing methods. While building a complete 3D dataset is tedious and expensive, abundant 2D in-the-wild data is already publicly available. As a consequence, there is a growing interest in the computer vision community to design efficient techniques that use the unsupervised learning strategy, which does not require any ground truth 3D data. Such methods can be trained with only natural 2D images of humans. In this paper we propose an unsupervised method for estimating 3D human pose in videos. The standard approach for unsupervised learning is to use the Generative Adversarial Network (GAN) framework. To improve the performance of 3D human pose estimation in videos, we propose a new GAN network that enforces body consistency over frames in a video. We evaluate the efficiency of our proposed method on a public 3D human body dataset.
引用
收藏
页码:5959 / 5964
页数:6
相关论文
共 50 条
  • [41] Cross View Fusion for 3D Human Pose Estimation
    Qiu, Haibo
    Wang, Chunyu
    Wang, Jingdong
    Wang, Naiyan
    Zeng, Wenjun
    2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, : 4341 - 4350
  • [42] MORE: simultaneous multi-view 3D object recognition and pose estimation
    Parisotto, Tommaso
    Mukherjee, Subhaditya
    Kasaei, Hamidreza
    INTELLIGENT SERVICE ROBOTICS, 2023, 16 (04) : 497 - 508
  • [43] Multi-Person Absolute 3D Pose and Shape Estimation from Video
    Zhang, Kaifu
    Li, Yihui
    Guan, Yisheng
    Xi, Ning
    INTELLIGENT ROBOTICS AND APPLICATIONS, ICIRA 2021, PT III, 2021, 13015 : 189 - 200
  • [44] Unsupervised universal hierarchical multi-person 3D pose estimation for natural scenes
    Renshu Gu
    Zhongyu Jiang
    Gaoang Wang
    Kevin McQuade
    Jenq-Neng Hwang
    Multimedia Tools and Applications, 2022, 81 : 32883 - 32906
  • [45] Unsupervised universal hierarchical multi-person 3D pose estimation for natural scenes
    Gu, Renshu
    Jiang, Zhongyu
    Wang, Gaoang
    McQuade, Kevin
    Hwang, Jenq-Neng
    MULTIMEDIA TOOLS AND APPLICATIONS, 2022, 81 (23) : 32883 - 32906
  • [46] A Multi-scale Recalibrated Approach for 3D Human Pose Estimation
    Xie, Ziwei
    Xia, Hailun
    Feng, Chunyan
    ADVANCES IN KNOWLEDGE DISCOVERY AND DATA MINING, PAKDD 2019, PT III, 2019, 11441 : 400 - 411
  • [47] 3D human pose estimation with multi-hypotheses gated transformer
    Dong, Xiena
    Zhang, Jian
    Yu, Jun
    Yu, Ting
    MULTIMEDIA SYSTEMS, 2024, 30 (06)
  • [48] 3D Human Pose Estimation Based on Multi-feature Extraction
    Ge, Senlin
    Yu, Huan
    Zhang, Yuanming
    Shi, Huitao
    Gao, Hao
    ARTIFICIAL INTELLIGENCE, CICAI 2022, PT III, 2022, 13606 : 570 - 581
  • [49] Enhanced 3D Pose Estimation in Multi-Person, Multi-View Scenarios through Unsupervised Domain Adaptation with Dropout Discriminator
    Deng, Junli
    Yao, Haoyuan
    Shi, Ping
    SENSORS, 2023, 23 (20)
  • [50] MHFormer: Multi-Hypothesis Transformer for 3D Human Pose Estimation
    Li, Wenhao
    Liu, Hong
    Tang, Hao
    Wang, Pichao
    Van Gool, Luc
    2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2022, : 13137 - 13146