Human pose estimation via Convolutional Part Heatmap Regression

TL;DR

This paper introduces a CNN cascade architecture for human pose estimation that effectively handles occlusions by combining part detection heatmaps with regression, improving accuracy on standard datasets.

Contribution

A novel CNN cascade architecture that learns part relationships and spatial context, robustly inferring pose even with severe occlusions.

Findings

Achieves top performance on MPII dataset.

Effectively handles occlusions through heatmap-guided regression.

Flexible integration with various CNN architectures.

Abstract

This paper is on human pose estimation using Convolutional Neural Networks. Our main contribution is a CNN cascaded architecture specifically designed for learning part relationships and spatial context, and robustly inferring pose even for the case of severe part occlusions. To this end, we propose a detection-followed-by-regression CNN cascade. The first part of our cascade outputs part detection heatmaps and the second part performs regression on these heatmaps. The benefits of the proposed architecture are multi-fold: It guides the network where to focus in the image and effectively encodes part constraints and context. More importantly, it can effectively cope with occlusions because part detection heatmaps for occluded parts provide low confidence scores which subsequently guide the regression part of our network to rely on contextual information in order to predict the location of these parts. Additionally, we show that the proposed cascade is flexible enough to readily allow the integration of various CNN architectures for both detection and regression, including recent ones based on residual learning. Finally, we illustrate that our cascade achieves top performance on the MPII and LSP data sets. Code can be downloaded from http://www.cs.nott.ac.uk/~psxab5/