Show simple item record

dc.contributor.authorCen, Feng
dc.contributor.authorWang, Guanghui
dc.date.accessioned2021-01-13T15:30:02Z
dc.date.available2021-01-13T15:30:02Z
dc.date.issued2019-02-25
dc.identifier.citationF. Cen and G. Wang, "Dictionary Representation of Deep Features for Occlusion-Robust Face Recognition," in IEEE Access, vol. 7, pp. 26595-26605, 2019. doi: 10.1109/ACCESS.2019.2901376en_US
dc.identifier.urihttp://hdl.handle.net/1808/31119
dc.description.abstractDeep learning has achieved exciting results in face recognition; however, the accuracy is still unsatisfying for occluded faces. To improve the robustness for occluded faces, this paper proposes a novel deep dictionary representation-based classification scheme, where a convolutional neural network is employed as the feature extractor and followed by a dictionary to linearly code the extracted deep features. The dictionary is composed by a gallery part consisting of the deep features of the training samples and an auxiliary part consisting of the mapping vectors acquired from the subjects either inside or outside the training set and associated with the occlusion patterns of the testing face samples. A squared Euclidean norm is used to regularize the coding coefficients. The proposed scheme is computationally efficient and is robust to large contiguous occlusion. In addition, the proposed scheme is generic for both the occluded and non-occluded face images and works with a single training sample per subject. The extensive experimental evaluations demonstrate the superior performance of the proposed approach over other state-of-the-art algorithms.en_US
dc.publisherInstitute of Electrical and Electronics Engineersen_US
dc.rightsCopyright © 2019, IEEE.en_US
dc.subjectFace recognitionen_US
dc.subjectConvolutional neural networken_US
dc.subjectOcclusion-robusten_US
dc.subjectDeep learningen_US
dc.subjectDictionary representationen_US
dc.titleDictionary Representation of Deep Features for Occlusion-Robust Face Recognitionen_US
dc.typeArticleen_US
kusw.kuauthorWang, Guanghui
kusw.kudepartmentElectrical Engineering & Computer Scienceen_US
dc.identifier.doi10.1109/ACCESS.2019.2901376en_US
dc.identifier.orcidhttps://orcid.org/0000-0002-0825-385Xen_US
dc.identifier.orcidhttps://orcid.org/0000-0003-3182-104Xen_US
kusw.oaversionScholarly/refereed, publisher versionen_US
kusw.oapolicyThis item meets KU Open Access policy criteria.en_US
dc.rights.accessrightsopenAccessen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record