Decanus to Legatus: Synthetic Training for 2D-3D Human Pose Lifting
Lists
Only you see this
Abstract
3D human pose estimation is a challenging task because of the difficulty to acquire ground-truth data outside of controlled environments. A number of further issues have been hindering progress in building a universal and robust model for this task, including domain gaps between different datasets, unseen actions between train and test datasets, various hardware settings and high cost of annotation, etc. In this paper, we propose an algorithm to generate infinite 3D synthetic human poses (Legatus) from a 3D pose distribution based on 10 initial handcrafted 3D poses (Decanus) during the training of a 2D to 3D human pose lifter neural network. Our results show that we can achieve 3D pose estimation performance comparable to methods using real data from specialized datasets but in a zero-shot setup, showing the generalization potential of our framework.
BibTeX
@inproceedings{zhu2022decanus,
author = {Yue Zhu and David Picard},
title = {{Decanus to Legatus: Synthetic Training for 2D-3D Human Pose Lifting}},
booktitle = {ACCV},
year = {2022},
url = {https://openreview.net/forum?id=xi0w8wBRJpo},
}| Lifting 2D Human Pose to 3D with Domain Adapted 3D Body Concept | arXiv · 2021 |
| Self-supervised 3D Human Pose Estimation from a Single Image | Workshops @ CVPR · 2023 |
| Cascaded Deep Monocular 3D Human Pose Estimation With Evolutionary Training Data | arXiv · 2020 |
| Lifting Motion to the 3D World via 2D Diffusion | CVPR · 2025 |
| Can 3D Pose Be Learned from 2D Projections Alone? | Workshops @ ECCV · 2018 |
| Unsupervised 3D Pose Estimation With Geometric Self-Supervision. | arXiv · 2019 |