Deep Fully-Connected Part-Based Models for Human Pose Estimation

Rodrigo de Bem, Anurag Arnab, Stuart Golodetz, Michael Sapienza, Philip Torr
Proceedings of The 10th Asian Conference on Machine Learning, PMLR 95:327-342, 2018.

Abstract

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.

Cite this Paper


BibTeX
@InProceedings{pmlr-v95-de-bem18a, title = {Deep Fully-Connected Part-Based Models for Human Pose Estimation}, author = {{de Bem}, Rodrigo and Arnab, Anurag and Golodetz, Stuart and Sapienza, Michael and Torr, Philip}, booktitle = {Proceedings of The 10th Asian Conference on Machine Learning}, pages = {327--342}, year = {2018}, editor = {Zhu, Jun and Takeuchi, Ichiro}, volume = {95}, series = {Proceedings of Machine Learning Research}, month = {14--16 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v95/de-bem18a/de-bem18a.pdf}, url = {https://proceedings.mlr.press/v95/de-bem18a.html}, abstract = {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.} }
Endnote
%0 Conference Paper %T Deep Fully-Connected Part-Based Models for Human Pose Estimation %A Rodrigo de Bem %A Anurag Arnab %A Stuart Golodetz %A Michael Sapienza %A Philip Torr %B Proceedings of The 10th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jun Zhu %E Ichiro Takeuchi %F pmlr-v95-de-bem18a %I PMLR %P 327--342 %U https://proceedings.mlr.press/v95/de-bem18a.html %V 95 %X 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.
APA
de Bem, R., Arnab, A., Golodetz, S., Sapienza, M. & Torr, P.. (2018). Deep Fully-Connected Part-Based Models for Human Pose Estimation. Proceedings of The 10th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 95:327-342 Available from https://proceedings.mlr.press/v95/de-bem18a.html.

Related Material