Unsupervised 3D Human Pose Estimation via Conditional Multi-view Ancestral Sampling
This work addresses the problem of 3D human pose estimation without 3D supervision, particularly for extreme poses where 3D data is scarce.
The authors propose an unsupervised 3D human pose estimation method using 2D diffusion priors from motion diffusion models, achieving better cross-domain performance on the Yoga dataset compared to supervised and unsupervised baselines.
We propose a method of estimating a 3D human pose from a single view without 3D supervision. The key to our method is to leverage the 2D diffusion priors of motion diffusion models (MDMs) pre-trained on large 2D human pose datasets. Specifically, we extend multi-view ancestral sampling of diffusion models to the task of 2D-3D lifting of human pose. To this end, we newly propose a conditional multi-view ancestral sampling (cMAS) that optimizes the 3D pose such that its multi-view projections follow the manifold in 2D MDM noise space, while conditioning the 3D pose to match the given 2D poses and anatomical constraints of humans. Experiments on the Yoga dataset demonstrate that our method achieves better cross-domain performance compared to state-of-the-art supervised and unsupervised 3D pose estimation methods, including extreme human poses where 3D supervision is unavailable. Code is available at: https://github.com/asaa0001/c-MAS.