Human 3D keypoints via spatial uncertainty modeling
This work addresses the problem of accurate 3D human keypoint estimation for researchers and applications in computer vision, offering a solution that reduces the dependency on expensive 3D annotations.
This paper proposes a method for 3D human keypoint estimation by directly modeling spatial uncertainty using robust statistics. The technique operates without 3D ground truth labels, relying only on 2D image-level keypoints, and achieves near state-of-the-art performance on the Human3.6m dataset.
We introduce a technique for 3D human keypoint estimation that directly models the notion of spatial uncertainty of a keypoint. Our technique employs a principled approach to modelling spatial uncertainty inspired from techniques in robust statistics. Furthermore, our pipeline requires no 3D ground truth labels, relying instead on (possibly noisy) 2D image-level keypoints. Our method achieves near state-of-the-art performance on Human3.6m while being efficient to evaluate and straightforward to