Deeply learning the messages in message passing inference

Simple item page
dc.contributor.authorLin, G.
dc.contributor.authorShen, C.
dc.contributor.authorReid, I.
dc.contributor.authorVan Den Hengel, A.
dc.contributor.conference29th Annual Conference on Neural Information Processing Systems 2015 (NIPS 2015) (7 Dec 2015 - 12 Dec 2015 : Montreal)
dc.contributor.editorCortes, C.
dc.contributor.editorLawrence, N.
dc.contributor.editorLee, D.
dc.contributor.editorSugiyama, M.
dc.contributor.editorGarnett, R.
dc.date.issued2015
dc.description.abstractDeep structured output learning shows great promise in tasks like semantic image segmentation. We proffer a new, efficient deep structured model learning scheme, in which we show how deep Convolutional Neural Networks (CNNs) can be used to directly estimate the messages in message passing inference for structured prediction with Conditional Random Fields CRFs). With such CNN message estimators, we obviate the need to learn or evaluate potential functions for message calculation. This confers significant efficiency for learning, since otherwise when performing structured learning for a CRF with CNN potentials it is necessary to undertake expensive inference for every stochastic gradient iteration. The network output dimension of message estimators is the same as the number of classes, rather than exponentially growing in the order of the potentials. Hence it is more scalable for cases that a large number of classes are involved. We apply our method to semantic image segmentation and achieve impressive performance, which demonstrates the effectiveness and usefulness of our CNN message learning method.
dc.description.statementofresponsibilityGuosheng Lin, Chunhua Shen, Ian Reid, Anton van den Hengel
dc.identifier.citationAdvances in neural information processing systems, 2015 / Cortes, C., Lawrence, N., Lee, D., Sugiyama, M., Garnett, R. (ed./s), vol.2015-January, pp.361-369
dc.identifier.isbn9781510825024
dc.identifier.issn1049-5258
dc.identifier.orcidReid, I. [0000-0001-7790-6423]
dc.identifier.orcidVan Den Hengel, A. [0000-0003-3027-8364]
dc.identifier.urihttp://hdl.handle.net/2440/115996
dc.language.isoen
dc.publisherNeural Information Processing Systems
dc.relation.granthttp://purl.org/au-research/grants/arc/CE140100016
dc.relation.granthttp://purl.org/au-research/grants/arc/FL130100102
dc.relation.ispartofseriesAdvances in Neural Information Processing Systems
dc.rightsCopyright © (2015) by Neural Information Processing Systems All rights reserved.
dc.titleDeeply learning the messages in message passing inference
dc.typeConference paper
pubs.publication-statusPublished

Files

Collections

Computer Science publications
Australian Institute for Machine Learning publications