Deep Convolutional Networks for Marker-less Human Pose Estimation from Multiple Views

Trumble, Matthew, Gilbert, Andrew, Hilton, Adrian and Collomosse, John (2016) Deep Convolutional Networks for Marker-less Human Pose Estimation from Multiple Views In: CVMP 2016. The 13th European Conference on Visual Media Production, 2016-12-12 - 2016-12-13, London, UK.

	Text CVMP_NoCopyright.pdf - Accepted version Manuscript Restricted to Repository staff only Available under License : See the attached licence file. Download (5MB)
Preview	Text (licence) SRI_deposit_agreement.pdf Available under License : See the attached licence file. Download (33kB) | Preview

Abstract

We propose a human performance capture system employing convolutional neural networks (CNN) to estimate human pose from a volumetric representation of a performer derived from multiple view-point video (MVV).We compare direct CNN pose regression to the performance of an affine invariant pose descriptor learned by a CNN through a classification task. A non-linear manifold embedding is learned between the descriptor and articulated pose spaces, enabling regression of pose from the source MVV. The results are evaluated against ground truth pose data captured using a Vicon marker-based system and demonstrate good generalisation over a range of human poses, providing a system that requires no special suit to be worn by the performer.

Item Type:	Conference or Workshop Item (Conference Paper)
Subjects :	Electronic Engineering
Divisions :	Faculty of Engineering and Physical Sciences > Electronic Engineering > Centre for Vision Speech and Signal Processing
Authors :	Name Email ORCID Trumble, Matthew Gilbert, Andrew A.Gilbert@surrey.ac.uk Hilton, Adrian A.Hilton@surrey.ac.uk Collomosse, John J.Collomosse@surrey.ac.uk
Date :	2016
DOI :	10.1145/2998559.2998565
Copyright Disclaimer :	© 2016 ACM
Uncontrolled Keywords :	Deep Learning; Pose Estimation; Multiple Viewpoint Video
Depositing User :	Symplectic Elements
Date Deposited :	08 Nov 2016 15:40
Last Modified :	17 Apr 2019 14:14
URI:	http://epubs.surrey.ac.uk/id/eprint/812671

Actions (login required)

View Item

Downloads