ProPLIKS: Probablistic 3D human body pose estimation

TL;DR

ProPLIKS introduces a probabilistic 3D human pose estimation method using normalizing flows on SO(3), effectively handling pose uncertainty and ambiguity, and outperforming existing techniques on RGB and X-Ray datasets.

Contribution

The paper presents a novel probabilistic framework employing normalizing flows on SO(3) for 3D human pose estimation, addressing discontinuities and ambiguities in pose reconstruction.

Findings

Outperforms existing pose estimation methods.

Effective on RGB images and X-Ray datasets.

Handles pose uncertainty and multi-view integration.

Abstract

We present a novel approach for 3D human pose estimation by employing probabilistic modeling. This approach leverages the advantages of normalizing flows in non-Euclidean geometries to address uncertain poses. Specifically, our method employs normalizing flow tailored to the SO(3) rotational group, incorporating a coupling mechanism based on the M\"obius transformation. This enables the framework to accurately represent any distribution on SO(3), effectively addressing issues related to discontinuities. Additionally, we reinterpret the challenge of reconstructing 3D human figures from 2D pixel-aligned inputs as the task of mapping these inputs to a range of probable poses. This perspective acknowledges the intrinsic ambiguity of the task and facilitates a straightforward integration method for multi-view scenarios. The combination of these strategies showcases the effectiveness of probabilistic models in complex scenarios for human pose estimation techniques. Our approach notably surpasses existing methods in the field of pose estimation. We also validate our methodology on human pose estimation from RGB images as well as medical X-Ray datasets.