Publication:
Identifying visual attributes for object recognition from text and taxonomy

dc.contributor.coauthorTırkaz, Çağlar
dc.contributor.coauthorEisenstein, Jacob
dc.contributor.coauthorYanıkoğlu, Berrin
dc.contributor.departmentDepartment of Computer Engineering
dc.contributor.kuauthorSezgin, Tevfik Metin
dc.contributor.schoolcollegeinstituteCollege of Engineering
dc.date.accessioned2024-11-09T22:53:39Z
dc.date.issued2015
dc.description.abstractAttributes of objects such as "square", "metallic", and "red" allow a way for humans to explain or discriminate object categories. These attributes also provide a useful intermediate representation for object recognition, including support for zero-shot learning from textual descriptions of object appearance. However, manual selection of relevant attributes among thousands of potential candidates is labor intensive. Hence, there is increasing interest in mining attributes for object recognition. In this paper, we introduce two novel techniques for nominating attributes and a method for assessing the suitability of candidate attributes for object recognition. The first technique for attribute nomination estimates attribute qualities based on their ability to discriminate objects at multiple levels of the taxonomy. The second technique leverages the linguistic concept of distributional similarity to further refine the estimated qualities. Attribute nomination is followed by our attribute assessment procedure, which assesses the quality of the candidate attributes based on their performance in object recognition. Our evaluations demonstrate that both taxonomy and distributional similarity serve as useful sources of information for attribute nomination, and our methods can effectively exploit them. We use the mined attributes in supervised and zero-shot learning settings to show the utility of the selected attributes in object recognition. Our experimental results show that in the supervised case we can improve on a state of the art classifier while in the zero-shot scenario we make accurate predictions outperforming previous automated techniques.
dc.description.indexedbyWOS
dc.description.indexedbyScopus
dc.description.openaccessNO
dc.description.publisherscopeInternational
dc.description.sponsoredbyTubitakEuN/A
dc.description.volume137
dc.identifier.doi10.1016/j.cviu.2015.02.006
dc.identifier.eissn1090-235X
dc.identifier.issn1077-3142
dc.identifier.quartileQ2
dc.identifier.scopus2-s2.0-84930539325
dc.identifier.urihttps://doi.org/10.1016/j.cviu.2015.02.006
dc.identifier.urihttps://hdl.handle.net/20.500.14288/7233
dc.identifier.wos356466800002
dc.keywordsObject recognition
dc.keywordsZero-shot learning
dc.keywordsAttribute mining
dc.keywordsAttribute-based classification
dc.language.isoeng
dc.publisherAcademic Press Inc Elsevier Science
dc.relation.ispartofComputer Vision and Image Understanding
dc.subjectComputer science
dc.subjectArtificial intelligence
dc.subjectEngineering, electrical electronic
dc.titleIdentifying visual attributes for object recognition from text and taxonomy
dc.typeJournal Article
dspace.entity.typePublication
local.contributor.kuauthorSezgin, Tevfik Metin
local.publication.orgunit1College of Engineering
local.publication.orgunit2Department of Computer Engineering
relation.isOrgUnitOfPublication89352e43-bf09-4ef4-82f6-6f9d0174ebae
relation.isOrgUnitOfPublication.latestForDiscovery89352e43-bf09-4ef4-82f6-6f9d0174ebae
relation.isParentOrgUnitOfPublication8e756b23-2d4a-4ce8-b1b3-62c794a8c164
relation.isParentOrgUnitOfPublication.latestForDiscovery8e756b23-2d4a-4ce8-b1b3-62c794a8c164

Files