PONet: Robust 3D Human Pose Estimation via Learning Orientations Only

TL;DR

PONet introduces a novel approach for 3D human pose estimation that learns orientations directly, making it more robust to occlusions and missing keypoints, and outperforming existing methods especially in challenging scenarios.

Contribution

The paper proposes PONet, a method that estimates 3D human pose using orientations only, bypassing the fragile keypoint detection step and improving robustness in occluded or truncated images.

Findings

Achieves state-of-the-art results in ideal conditions.

Significantly outperforms existing methods in occlusion scenarios.

Reduces dependency on 2D keypoint detectors.

Abstract

Conventional 3D human pose estimation relies on first detecting 2D body keypoints and then solving the 2D to 3D correspondence problem.Despite the promising results, this learning paradigm is highly dependent on the quality of the 2D keypoint detector, which is inevitably fragile to occlusions and out-of-image absences.In this paper,we propose a novel Pose Orientation Net (PONet) that is able to robustly estimate 3D pose by learning orientations only, hence bypassing the error-prone keypoint detector in the absence of image evidence. For images with partially invisible limbs, PONet estimates the 3D orientation of these limbs by taking advantage of the local image evidence to recover the 3D pose.Moreover, PONet is competent to infer full 3D poses even from images with completely invisible limbs, by exploiting the orientation correlation between visible limbs to complement the estimated poses,further improving the robustness of 3D pose estimation.We evaluate our method on multiple datasets, including Human3.6M, MPII, MPI-INF-3DHP, and 3DPW. Our method achieves results on par with state-of-the-art techniques in ideal settings, yet significantly eliminates the dependency on keypoint detectors and the corresponding computation burden. In highly challenging scenarios, such as truncation and erasing, our method performs very robustly and yields much superior results as compared to state of the art,demonstrating its potential for real-world applications.