Bayesian semantic instance segmentation in open set world

Simple item page
dc.contributor.authorPham, T.
dc.contributor.authorVijay Kumar, B.
dc.contributor.authorDo, T.
dc.contributor.authorCarneiro, G.
dc.contributor.authorReid, I.
dc.contributor.conferenceEuropean Conference on Computer Vision (ECCV) (8 Sep 2018 - 14 Sep 2018 : Munich)
dc.contributor.editorFerrari, V.
dc.contributor.editorHebert, M.
dc.contributor.editorSminchisescu, C.
dc.contributor.editorWeiss, Y.
dc.date.issued2018
dc.description.abstractThis paper addresses the semantic instance segmentation task in the open-set conditions, where input images can contain known and unknown object classes. The training process of existing semantic instance segmentation methods requires annotation masks for all object instances, which is expensive to acquire or even infeasible in some realistic scenarios, where the number of categories may increase boundlessly. In this paper, we present a novel open-set semantic instance segmentation approach capable of segmenting all known and unknown object classes in images, based on the output of an object detector trained on known object classes. We formulate the problem using a Bayesian framework, where the posterior distribution is approximated with a simulated annealing optimization equipped with an efficient image partition sampler. We show empirically that our method is competitive with state-of-the-art supervised methods on known classes, but also performs well on unknown classes when compared with unsupervised methods.
dc.identifier.citationLecture Notes in Artificial Intelligence, 2018 / Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (ed./s), vol.11214 LNCS, pp.3-18
dc.identifier.doi10.1007/978-3-030-01249-6_1
dc.identifier.isbn9783030012458
dc.identifier.issn0302-9743
dc.identifier.issn1611-3349
dc.identifier.orcidCarneiro, G. [0000-0002-5571-6220]
dc.identifier.orcidReid, I. [0000-0001-7790-6423]
dc.identifier.urihttp://hdl.handle.net/2440/116990
dc.language.isoen
dc.publisherSpringer
dc.relation.ispartofseriesLecture Notes in Computer Science; 11205
dc.rights© Springer Nature Switzerland AG 2018
dc.source.urihttps://doi.org/10.1007/978-3-030-01249-6_1
dc.titleBayesian semantic instance segmentation in open set world
dc.typeConference paper
pubs.publication-statusPublished

Files

Collections

Computer Science publications
Australian Institute for Machine Learning publications