Perceiving Systems, Computer Vision

AGORA: Avatars in Geography Optimized for Regression Analysis

2021

Conference Paper

ps


While the accuracy of 3D human pose estimation from images has steadily improved on benchmark datasets, the best methods still fail in many real-world scenarios. This suggests that there is a domain gap between current datasets and common scenes containing people. To obtain ground-truth 3D pose, current datasets limit the complexity of clothing, environmental conditions, number of subjects, and occlusion. Moreover, current datasets evaluate sparse 3D joint locations corresponding to the major joints of the body, ignoring the hand pose and the face shape. To evaluate the current state-of-the-art methods on more challenging images, and to drive the field to address new problems, we introduce AGORA, a synthetic dataset with high realism and highly accurate ground truth. Here we use 4240 commercially-available, high-quality, textured human scans in diverse poses and natural clothing; this includes 257 scans of children. We create reference 3D poses and body shapes by fitting the SMPL-X body model (with face and hands) to the 3D scans, taking into account clothing. We create around 14K training and 3K test images by rendering between 5 and 15 people per image us- ing either image-based lighting or rendered 3D environments, taking care to make the images physically plausible and photoreal. In total, AGORA consists of 173K individual person crops. We evaluate existing state-of-the- art methods for 3D human pose estimation on this dataset. and find that most methods perform poorly on images of children. Hence, we extend the SMPL-X model to better capture the shape of children. Additionally, we fine- tune methods on AGORA and show improved performance on both AGORA and 3DPW, confirming the realism of the dataset. We provide all the registered 3D reference training data, rendered images, and a web-based evaluation site at https://agora.is.tue.mpg.de/.

Author(s): Priyanka Patel and Chun-Hao P. Huang and Joachim Tesch and David T. Hoffmann and Shashank Tripathi and Michael J. Black
Book Title: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2021)
Pages: 13463--13473
Year: 2021
Month: June
Publisher: IEEE

Department(s): Perceiving Systems
Bibtex Type: Conference Paper (inproceedings)
Paper Type: Conference

DOI: 10.1109/CVPR46437.2021.01326
Event Name: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2021)
Event Place: Virtual

Address: Piscataway, NJ
ISBN: 978-1-6654-4510-8
State: Published

Links: dataset
pdf
video
Video:

BibTex

@inproceedings{Patel:CVPR:2021,
  title = {{AGORA}: Avatars in Geography Optimized for Regression Analysis},
  author = {Patel, Priyanka and Huang, Chun-Hao P. and Tesch, Joachim and Hoffmann, David T. and Tripathi, Shashank and Black, Michael J.},
  booktitle = {2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2021)},
  pages = {13463--13473},
  publisher = {IEEE},
  address = {Piscataway, NJ},
  month = jun,
  year = {2021},
  doi = {10.1109/CVPR46437.2021.01326},
  month_numeric = {6}
}