Pose Prior Learner: Unsupervised Categorical Prior Learning for Pose Estimation

TL;DR

This paper introduces Pose Prior Learner (PPL), a self-supervised method that learns a general pose prior from images to improve pose estimation accuracy for both humans and animals, especially in occluded scenarios.

Contribution

PPL is a novel hierarchical memory-based approach that learns meaningful pose priors without human annotations, enhancing pose estimation through template transformation and iterative inference.

Findings

Outperforms baselines on human and animal pose datasets.

Effectively handles occluded images using learned pose priors.

Improves pose estimation accuracy via iterative refinement.

Abstract

A prior represents a set of beliefs or assumptions about a system, aiding inference and decision-making. In this paper, we introduce the challenge of unsupervised categorical prior learning in pose estimation, where AI models learn a general pose prior for an object category from images in a self-supervised manner. Although priors are effective in estimating pose, acquiring them can be difficult. We propose a novel method, named Pose Prior Learner (PPL), to learn a general pose prior for any object category. PPL uses a hierarchical memory to store compositional parts of prototypical poses, from which we distill a general pose prior. This prior improves pose estimation accuracy through template transformation and image reconstruction. PPL learns meaningful pose priors without any additional human annotations or interventions, outperforming competitive baselines on both human and animal pose estimation datasets. Notably, our experimental results reveal the effectiveness of PPL using learned prototypical poses for pose estimation on occluded images. Through iterative inference, PPL leverages the pose prior to refine estimated poses, regressing them to any prototypical poses stored in memory. Our code, model, and data are publicly available at: https://github.com/ZhangLab-DeepNeuroCogLab/Pose-Prior-Learner.