Deep fully-connected part-based models for human pose estimation

Conference item

We propose a 2D multi-level appearance representation of the human body in RGB images, spatially modelled using a fully-connected graphical model. The appearance model is based on a CNN body part detector, which uses shared features in a cascade architecture to simultaneously detect body parts with different levels of granularity. We use a fully-connected Conditional Random Field (CRF) as our spatial model, over which approximate inference is efficiently performed using the Mean-Field algorithm, implemented as a Recurrent Neural Network (RNN). The stronger visual support from body parts with different levels of granularity, along with the fully-connected pairwise spatial relations, which have their weights learnt by the model, improve the performance of the bottom-up part detector. We adopt an end-to-end training strategy to leverage the potential of both our appearance and spatial models, and achieve competitive results on the MPII and LSP datasets.

Authors: De Bem, R; Arnab, A; Golodetz, S; Sapienza, M; Torr, P

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Proceedings of Machine Learning Research
• Host title: 10th Asian Conference on Machine Learning (ACML 2018)
• Journal: Asian Conference on Machine Learning (ACML 2018)
• Publication date: 2018-11-04
• Acceptance date: 2018-09-20
• ISSN: 1938-7228
• Keywords: CRFs; deep learning; human pose estimation; fully-connected part-based models