Self-Supervised 3D Human Pose Estimation with Multiple-View Geometry

Bouazizi A, Wiederer J, Kressel U, Belagiannis V (2021)

Publication Type: Conference contribution

Publication year: 2021

Publisher: Institute of Electrical and Electronics Engineers Inc.

Conference Proceedings Title: Proceedings - 2021 16th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2021

Event location: Virtual, Jodhpur, IND

ISBN: 9781665431767

DOI: 10.1109/FG52635.2021.9667074

Abstract

We present a self-supervised learning algorithm for 3D human pose estimation of a single person based on a multiple-view camera system and 2D body pose estimates for each view. To train our model, represented by a deep neural network, we propose a four-loss function learning algorithm, which does not require any 2D or 3D body pose ground-truth. The proposed loss functions make use of the multiple-view geometry to reconstruct 3D body pose estimates and impose body pose constraints across the camera views. Our approach utilizes all available camera views during training, while the inference is single-view. In our evaluations, we show promising performance on Human3.6M and HumanEva benchmarks, while we also present a generalization study on MPI-INF-3DHP dataset, as well as several ablation results. Overall, we outperform all self-supervised learning methods and reach comparable results to supervised and weakly-supervised learning approaches. Our code and models are publicly available11Source Code: https://gi-thub.com/vru2020jpose_3D/.

Authors with CRIS profile

Vasileios Belagiannis

Involved external institutions

Daimler AG DE Germany (DE) Universität Ulm DE Germany (DE)

How to cite

APA:

Bouazizi, A., Wiederer, J., Kressel, U., & Belagiannis, V. (2021). Self-Supervised 3D Human Pose Estimation with Multiple-View Geometry. In Vitomir Struc, Marija Ivanovska (Eds.), Proceedings - 2021 16th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2021. Virtual, Jodhpur, IND: Institute of Electrical and Electronics Engineers Inc..

MLA:

Bouazizi, Arij, et al. "Self-Supervised 3D Human Pose Estimation with Multiple-View Geometry." Proceedings of the 16th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2021, Virtual, Jodhpur, IND Ed. Vitomir Struc, Marija Ivanovska, Institute of Electrical and Electronics Engineers Inc., 2021.

BibTeX: Download