3D-Aware Neural Body Fitting for Occlusion Robust 3D Human Pose Estimation

TL;DR

This paper introduces 3D-aware Neural Body Fitting (3DNBF), a novel analysis-by-synthesis approach that leverages a generative model of deep features for robust 3D human pose estimation, especially under occlusion.

Contribution

It proposes a generative model of deep features based on volumetric human representation and contrastive learning to improve occlusion robustness in 3D pose estimation.

Findings

Outperforms existing methods on occluded benchmarks

Achieves state-of-the-art performance on standard datasets

Demonstrates robustness to occlusion in 3D human pose estimation

Abstract

Regression-based methods for 3D human pose estimation directly predict the 3D pose parameters from a 2D image using deep networks. While achieving state-of-the-art performance on standard benchmarks, their performance degrades under occlusion. In contrast, optimization-based methods fit a parametric body model to 2D features in an iterative manner. The localized reconstruction loss can potentially make them robust to occlusion, but they suffer from the 2D-3D ambiguity. Motivated by the recent success of generative models in rigid object pose estimation, we propose 3D-aware Neural Body Fitting (3DNBF) - an approximate analysis-by-synthesis approach to 3D human pose estimation with SOTA performance and occlusion robustness. In particular, we propose a generative model of deep features based on a volumetric human representation with Gaussian ellipsoidal kernels emitting 3D pose-dependent feature vectors. The neural features are trained with contrastive learning to become 3D-aware and hence to overcome the 2D-3D ambiguity. Experiments show that 3DNBF outperforms other approaches on both occluded and standard benchmarks. Code is available at https://github.com/edz-o/3DNBF