Deep Hierarchical Human Semantic Parsing
Wenguan Wang,Tianfei Zhou,Jianbing Shen,Song-Chun Zhu
2020-01-01
Abstract:Human parsing is for pixel-wise human semantic understanding. As human bodies are hierarchically structured, how to model human structures is the central theme in this task. We start with analyzing three inference processes over the human hierarchy: direct inference (directly predicting human semantic parts using image information), bottom-up inference (assembling knowledge from constituent parts), and top-down inference (leveraging context from parent nodes). We then formulate the problem as a compositional neural information fusion (CNIF) framework, which assembles the information from the three inference processes in a conditional manner, i.e., considering the confidence of the sources. Based on CNIF, we further present a part-relation aware human parser (PRHP), which precisely describes three kinds of human part relations, i.e., decomposition, composition, and dependency, by three distinct relation networks. Expressive relation information can be captured by imposing the parameters in the relation networks to satisfy specific geometric characteristics of different relations. By assimilating generic message-passing networks with their edge-typed, convolutional counterparts, PRHP performs iterative reasoning over the human hierarchy. With these efforts, PRHP provides a more general and powerful form of CNIF, and lays the foundation for more sophisticated and flexible human relation patterns of reasoning. Experiments on five datasets demonstrate that our two human parsers outperform the state-of-the-arts in all cases.