Iterative Spectral Clustering for Unsupervised Object Localization

Show simple item record

dc.contributor.author Vora, Aditya
dc.contributor.author Raman, Shanmuganathan
dc.date.accessioned 2018-02-28T11:39:18Z
dc.date.available 2018-02-28T11:39:18Z
dc.date.issued 2018-04
dc.identifier.citation Vora, Aditya and Raman, Shanmuganathan, “Iterative spectral clustering for unsupervised object localization”, Pattern Recognition Letters, DOI: 10.1016/j.patrec.2018.02.012, vol.106, pp.27-32.Apr. 2018. en_US
dc.identifier.issn 0167-8655
dc.identifier.issn 1872-7344
dc.identifier.uri https://repository.iitgn.ac.in/handle/123456789/3492
dc.identifier.uri http://dx.doi.org/10.1016/j.patrec.2018.02.012
dc.description.abstract This paper addresses the problem of unsupervised object localization in an image. Unlike previous supervised and weakly supervised algorithms that require bounding box or image level annotations for training classifiers, we propose a simple yet effective technique for localization using iterative spectral clustering. This iterative spectral clustering approach along with appropriate cluster selection strategy in each iteration naturally helps in searching of object region in the image. In order to estimate the final localization window, we group the proposals obtained from the iterative spectral clustering step based on the perceptual similarity, and average the coordinates of the proposals from the top scoring groups. We benchmark our algorithm on challenging datasets like Object Discovery and PASCAL VOC 2007, achieving an average CorLoc percentage of 51% and 35% respectively which is comparable to various other weakly supervised algorithms despite being completely unsupervised. en_US
dc.description.statementofresponsibility Aditya Vora and Shanmuganathan Raman
dc.format.extent vol.106, pp.27-32.
dc.language.iso en en_US
dc.publisher Elsevier en_US
dc.subject Object Localization en_US
dc.subject Spectral Clustering en_US
dc.subject Unsupervised Localization en_US
dc.title Iterative Spectral Clustering for Unsupervised Object Localization en_US
dc.type Article en_US
dc.relation.journal Pattern Recognition Letters


Files in this item

Files Size Format View

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record

Search Digital Repository


Browse

My Account