Unsupervised Human Pose Estimation through Transforming Shape Templates

TL;DR

This paper introduces an unsupervised method for human pose estimation that uses learnable shape templates and deep features, capable of generalizing across different human shapes including infants.

Contribution

It presents a novel unsupervised approach using shape templates and deep features for pose estimation, applicable to diverse human body types without large annotated datasets.

Findings

Effective on datasets of adults and infants

Does not require large annotated datasets

Generalizes across different body shapes

Abstract

Human pose estimation is a major computer vision problem with applications ranging from augmented reality and video capture to surveillance and movement tracking. In the medical context, the latter may be an important biomarker for neurological impairments in infants. Whilst many methods exist, their application has been limited by the need for well annotated large datasets and the inability to generalize to humans of different shapes and body compositions, e.g. children and infants. In this paper we present a novel method for learning pose estimators for human adults and infants in an unsupervised fashion. We approach this as a learnable template matching problem facilitated by deep feature extractors. Human-interpretable landmarks are estimated by transforming a template consisting of predefined body parts that are characterized by 2D Gaussian distributions. Enforcing a connectivity prior guides our model to meaningful human shape representations. We demonstrate the effectiveness of our approach on two different datasets including adults and infants.