This repository provides the official PyTorch implementation of the HRNet (High-Resolution Representation Learning) architecture for human pose estimation, published at CVPR 2019. The model maintains high-resolution representations throughout the network to accurately detect body keypoints (such as joints) on humans in images, trained on COCO and MPII datasets. The HRNet architecture has also been extended to other vision tasks including semantic segmentation, object detection, and facial landmark detection.