Human pose estimation via multi-layer composite models

We introduce a hierarchical part-based approach for human pose estimation in static images. Our model is a multi-layer composite of tree-structured pictorial-structure models, each modeling human pose at a different scale and with a different graphical structure. At the highest level, the submodel acts as a person detector, while at the lowest level, the body is decomposed into a collection of many local parts. Edges between adjacent layers of the composite model encode cross-model constraints. This multi-layer composite model is able to relax the independence assumptions in tree-structured pictorial-structures models (which can create problems like double-counting image evidence), while still permitting efficient inference using dual-decomposition. We propose an optimization procedure for joint learning of the entire composite model. Our approach outperforms the state-of-the-art on four challenging datasets: Parse, UIUC Sport, Leeds Sport Pose and FLIC datasets. HighlightsWe propose a novel multi-layer graphical model for human pose estimation.We conduct a broad and detailed literature review on hierarchical models, multi-scale models, and mixture models for human pose estimation.We propose a novel design of dual decomposition approach for efficient inference applied on the proposed multi-layer composite model.We propose a joint discriminative learning framework using structured SVM.We performed systematic evaluation on three challenging state-of-art pose estimation datasets using our model against baseline approaches.