3DPW

3D human pose and mesh estimation, the task of recon-structing human pose in 3D space given a 2D image of the person, is an ill-posed problem, and many data-driven ap-proaches using deep learning were recently proposed (Liu, Kortylewski, and Yuille 2023; Gholami et al. 2022a; Li et al. 2022). A dataset covering the distribution of possible human poses, global orientation, appearance, and other attributes would be large and difficult to capture in practice.

Data and Resources

Cite this as

Mohsen Gholami, Rabab Ward, Z. Jane Wang (2024). Dataset: 3DPW. https://doi.org/10.57702/ftenjlve

DOI retrieved: December 2, 2024

Additional Info

Field Value
Created December 2, 2024
Last update December 2, 2024
Defined In https://doi.org/10.48550/arXiv.2405.07407
Citation
  • https://doi.org/10.48550/arXiv.2312.14915
  • https://doi.org/10.48550/arXiv.2405.19833
  • https://doi.org/10.48550/arXiv.2208.10652
  • https://doi.org/10.48550/arXiv.2012.09843
  • https://doi.org/10.48550/arXiv.2403.09050
  • https://doi.org/10.48550/arXiv.2309.04840
  • https://doi.org/10.48550/arXiv.2105.09711
  • https://doi.org/10.48550/arXiv.2304.04875
  • https://doi.org/10.48550/arXiv.2104.05015
  • https://doi.org/10.48550/arXiv.2007.11755
Author Mohsen Gholami
More Authors
Rabab Ward
Z. Jane Wang
Homepage https://www.3dpw.org/