Self-supervised Structure-Sensitive Learning for Human Parsing

Human parsing has recently attracted much research interest due to its enormous application potential. In this chapter, we introduce a new benchmark,“Look into Person (LIP),” that makes a significant advance in terms of scalability, diversity, and difficulty, a contribution that we feel is crucial for future developments in human-centric analysis. Furthermore, in contrast to the existing efforts to improve feature discriminative capability, we solve human parsing by exploring a novel self-supervised structure-sensitive learning approach that imposes human pose structures on the parsing results without requiring extra supervision. Our self-supervised learning framework can be injected into any advanced neural network to help incorporate rich high-level knowledge regarding human joints from a global perspective and improve the parsing results (©[2019] IEEE., Reprinted, with permission, from [1]).