DiffPose: Toward More Reliable 3D Pose Estimation

TL;DR

DiffPose introduces a novel diffusion-based framework for monocular 3D human pose estimation, effectively handling ambiguity and occlusion, and significantly outperforming existing methods on standard benchmarks.

Contribution

The paper proposes a diffusion process formulation for 3D pose estimation, incorporating pose-specific initialization and a Gaussian Mixture Model-based diffusion process.

Findings

Outperforms existing methods on Human3.6M and MPI-INF-3DHP benchmarks.

Effectively handles ambiguity and occlusion in pose estimation.

Demonstrates the potential of diffusion models for 3D pose tasks.

Abstract

Monocular 3D human pose estimation is quite challenging due to the inherent ambiguity and occlusion, which often lead to high uncertainty and indeterminacy. On the other hand, diffusion models have recently emerged as an effective tool for generating high-quality images from noise. Inspired by their capability, we explore a novel pose estimation framework (DiffPose) that formulates 3D pose estimation as a reverse diffusion process. We incorporate novel designs into our DiffPose to facilitate the diffusion process for 3D pose estimation: a pose-specific initialization of pose uncertainty distributions, a Gaussian Mixture Model-based forward diffusion process, and a context-conditioned reverse diffusion process. Our proposed DiffPose significantly outperforms existing methods on the widely used pose estimation benchmarks Human3.6M and MPI-INF-3DHP. Project page: https://gongjia0208.github.io/Diffpose/.