DiffuPose: Monocular 3D Human Pose Estimation via Denoising Diffusion Probabilistic Model

被引:7
|
作者
Choi, Jeongjun [1 ,2 ]
Shim, Dongseok [1 ]
Kim, H. Jin [1 ,2 ]
机构
[1] Seoul Natl Univ, Artificial Intelligence Inst AIIS, Seoul, South Korea
[2] Automat & Syst Res Inst ASRI, Seoul, South Korea
基金
新加坡国家研究基金会;
关键词
D O I
10.1109/IROS55552.2023.10342204
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Thanks to the development of 2D keypoint detectors, monocular 3D human pose estimation (HPE) via 2D-to-3D uplifting approaches have achieved remarkable improvements. Still, monocular 3D HPE is a challenging problem due to the inherent depth ambiguities and occlusions. To handle this problem, many previous works exploit temporal information to mitigate such difficulties. However, there are many real-world applications where frame sequences are not accessible. This paper focuses on reconstructing a 3D pose from a single 2D keypoint detection. Rather than exploiting temporal information, we alleviate the depth ambiguity by generating multiple 3D pose candidates which can be mapped to an identical 2D keypoint. We build a novel diffusion-based framework to effectively sample diverse 3D poses from an off-the-shelf 2D detector. By considering the correlation between human joints by replacing the conventional denoising U-Net with graph convolutional network, our approach accomplishes further performance improvements. We evaluate our method on the widely adopted Human3.6M and HumanEva-I datasets. Comprehensive experiments are conducted to prove the efficacy of the proposed method, and they confirm that our model outperforms state-of-the-art multi-hypothesis 3D HPE methods.
引用
收藏
页码:3773 / 3780
页数:8
相关论文
共 50 条
  • [1] Probabilistic Monocular 3D Human Pose Estimation with Normalizing Flows
    Wehrbein, Tom
    Rudolph, Marco
    Rosenhahn, Bodo
    Wandt, Bastian
    2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021), 2021, : 11179 - 11188
  • [2] A survey on monocular 3D human pose estimation
    Ji X.
    Fang Q.
    Dong J.
    Shuai Q.
    Jiang W.
    Zhou X.
    Virtual Reality and Intelligent Hardware, 2020, 2 (06): : 471 - 500
  • [3] MONOCULAR 3D HUMAN POSE ESTIMATION BY CLASSIFICATION
    Greif, Thomas
    Lienhart, Rainer
    Sengupta, Debabrata
    2011 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO (ICME), 2011,
  • [4] Adapted human pose: monocular 3D human pose estimation with zero real 3D pose data
    Liu, Shuangjun
    Sehgal, Naveen
    Ostadabbas, Sarah
    APPLIED INTELLIGENCE, 2022, 52 (12) : 14491 - 14506
  • [5] Adapted human pose: monocular 3D human pose estimation with zero real 3D pose data
    Shuangjun Liu
    Naveen Sehgal
    Sarah Ostadabbas
    Applied Intelligence, 2022, 52 : 14491 - 14506
  • [6] Monocular 3D Pose Estimation via Pose Grammar and Data Augmentation
    Xu, Yuanlu
    Wang, Wenguan
    Liu, Tengyu
    Liu, Xiaobai
    Xie, Jianwen
    Zhu, Song-Chun
    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2022, 44 (10) : 6327 - 6344
  • [7] Generalizing Monocular 3D Human Pose Estimation in the Wild
    Wang, Luyang
    Chen, Yan
    Guo, Zhenhua
    Qian, Keyuan
    Lin, Mude
    Li, Hongsheng
    Ren, Jimmy S.
    2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCVW), 2019, : 4024 - 4033
  • [8] Monocular vehicle pose estimation based on 3D model
    Xu L.-Z.
    Fu Q.-W.
    Tao W.
    Zhao H.
    Guangxue Jingmi Gongcheng/Optics and Precision Engineering, 2021, 29 (06): : 1346 - 1355
  • [9] Denoising Diffusion for 3D Hand Pose Estimation from Images
    Ivashechkin, Maksym
    Mendez, Oscar
    Bowden, Richard
    2023 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS, ICCVW, 2023, : 3128 - 3137
  • [10] Joint Optimization of the 3D Model and 6D Pose for Monocular Pose Estimation
    Guo, Liangchao
    Chen, Lin
    Wang, Qiufu
    Zhang, Zhuo
    Sun, Xiaoliang
    DRONES, 2024, 8 (11)