Deep High-Resolution Representation Learning for Human Pose Estimation

TL;DR

This paper introduces a high-resolution network architecture for human pose estimation that maintains high-resolution representations throughout the process, leading to more accurate and spatially precise keypoint detection.

Contribution

It proposes a novel high-resolution network that preserves high-resolution features via multi-scale fusion, improving pose estimation accuracy over existing methods.

Findings

Achieves superior results on COCO and MPII datasets.

Maintains high-resolution features for better spatial accuracy.

Demonstrates effectiveness of multi-scale fusion in pose estimation.

Abstract

This is an official pytorch implementation of Deep High-Resolution Representation Learning for Human Pose Estimation. In this work, we are interested in the human pose estimation problem with a focus on learning reliable high-resolution representations. Most existing methods recover high-resolution representations from low-resolution representations produced by a high-to-low resolution network. Instead, our proposed network maintains high-resolution representations through the whole process. We start from a high-resolution subnetwork as the first stage, gradually add high-to-low resolution subnetworks one by one to form more stages, and connect the mutli-resolution subnetworks in parallel. We conduct repeated multi-scale fusions such that each of the high-to-low resolution representations receives information from other parallel representations over and over, leading to rich high-resolution representations. As a result, the predicted keypoint heatmap is potentially more accurate and spatially more precise. We empirically demonstrate the effectiveness of our network through the superior pose estimation results over two benchmark datasets: the COCO keypoint detection dataset and the MPII Human Pose dataset. The code and models have been publicly available at \url{https://github.com/leoxiaobin/deep-high-resolution-net.pytorch}.