An adversarial human pose estimation network injected with graph structure. (July 2021)

• Record Type: Journal Article
• Authors: Tian, Lei; Wang, Peng; Liang, Guoqiang; Shen, Chunhua
• Is Part Of: Pattern recognition. Volume 115(2021)
• Publication Date: 2021-07
• Subjects: Human pose estimation -- Cascade feature network -- Graph structure network -- Generative adversarial network; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2021.107863
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We inject the graph structure into a GAN-based human pose estimation model to capture the relationship of body joints. Because the graph based discriminator is not needed during testing, the complexity of our pose estimation model does not increase. Experimental results on three public benchmark datasets show the effectiveness of our method. Abstract: Because of the invisible human keypoints in images caused by illumination, occlusion and overlap, it is likely to produce unreasonable human pose prediction for most of the current human pose estimation methods. In this paper, we design a novel generative adversarial network (GAN) to improve the localization accuracy of visible joints when some joints are invisible. The network consists of two simple but efficient modules, i.e ., Cascade Feature Network (CFN) and Graph Structure Network (GSN). First, the CFN utilizes the prediction maps from the previous stages to guide the prediction maps in the next stage to produce accurate human pose. Second, the GSN is designed to contribute to the localization of invisible joints by passing message among different joints. According to GAN, if the prediction pose produced by the generator G cannot be distinguished by the discriminator D, the generator network G has successfully obtained the underlying dependence of human joints. We conduct experiments on three widely used human pose estimation benchmark datasets, i.e ., LSP, MPII and COCO, whose results show the effectiveness of our proposed framework.