Cascaded Pyramid Network for Multi-person Pose Estimation

TLDR: A novel network structure called Cascaded Pyramid Network (CPN) is presented which targets to relieve the problem from these "hard" keypoints, with state-of-art results on the COCO keypoint benchmark, with average precision at 73.0.

Abstract: The topic of multi-person pose estimation has been largely improved recently, especially with the development of convolutional neural network. However, there still exist a lot of challenging cases, such as occluded keypoints, invisible keypoints and complex background, which cannot be well addressed. In this paper, we present a novel network structure called Cascaded Pyramid Network (CPN) which targets to relieve the problem from these "hard" keypoints. More specifically, our algorithm includes two stages: GlobalNet and RefineNet. GlobalNet is a feature pyramid network which can successfully localize the "simple" keypoints like eyes and hands but may fail to precisely recognize the occluded or invisible keypoints. Our RefineNet tries explicitly handling the "hard" keypoints by integrating all levels of feature representations from the GlobalNet together with an online hard keypoint mining loss. In general, to address the multi-person pose estimation problem, a top-down pipeline is adopted to first generate a set of human bounding boxes based on a detector, followed by our CPN for keypoint localization in each human bounding box. Based on the proposed algorithm, we achieve state-of-art results on the COCO keypoint benchmark, with average precision at 73.0 on the COCO test-dev dataset and 72.1 on the COCO test-challenge dataset, which is a 19% relative improvement compared with 60.5 from the COCO 2016 keypoint challenge. Code1 and the detection results for person used will be publicly available for further research.