Evaluating Spatial Configuration Constrained CNNs for Localizing Facial and Body Pose Landmarks

ResearchSpace/Manakin Repository

Show simple item record

dc.contributor.author Payer, C en
dc.contributor.author Štern, D en
dc.contributor.author Urschler, Martin en
dc.date.accessioned 2020-04-15T04:40:38Z en
dc.date.issued 2019-12-01 en
dc.identifier.isbn 9781728141879 en
dc.identifier.issn 2151-2191 en
dc.identifier.uri http://hdl.handle.net/2292/50396 en
dc.description.abstract © 2019 IEEE. Landmark localization is a widely used task required in medical image analysis and computer vision applications. Formulated in a heatmap regression framework, we have recently proposed a CNN architecture that learns on its own to split the localization task into two simpler sub-problems, dedicating one component to locally accurate but ambiguous predictions, while the other component improves robustness by incorporating the spatial configuration of landmarks to remove ambiguities. We learn this simplification in our SpatialConfiguration-Net (SCN) by multiplying the heatmap predictions of its two components and by training the network in and end-to-end manner, thus achieving regularization similar to e.g. a hand-crafted Markov Random Field model. While we have previously shown localization results solely on data from 2D and 3D medical imaging modalities, in this work our aim is to study the generalization capabilities of our SpatialConfiguration-Net to computer vision problems. Therefore, we evaluate our performance both in terms of accuracy and robustness on a facial alignment task, where we improve upon the state-of-the-art methods, as well as on a human body pose estimation task, where we demonstrate results in line with the recent state-of-the-art. en
dc.relation.ispartofseries International Conference Image and Vision Computing New Zealand en
dc.rights Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated. Previously published items are made available in accordance with the copyright policy of the publisher. en
dc.rights.uri https://researchspace.auckland.ac.nz/docs/uoa-docs/rights.htm en
dc.title Evaluating Spatial Configuration Constrained CNNs for Localizing Facial and Body Pose Landmarks en
dc.type Conference Item en
dc.identifier.doi 10.1109/IVCNZ48456.2019.8961000 en
pubs.volume 2019-December en
dc.rights.holder Copyright: The author en
pubs.publication-status Published en
dc.rights.accessrights http://purl.org/eprint/accessRights/RestrictedAccess en
pubs.elements-id 793775 en
pubs.org-id Science en
pubs.org-id School of Computer Science en
dc.identifier.eissn 2151-2205 en

Full text options

Full text for this item is not available in ResearchSpace.

Find Full text

This item appears in the following Collection(s)

Show simple item record


Search ResearchSpace

Advanced Search