Show simple item record

dc.contributor.authorWang, Chen
dc.contributor.authorWang, Xiang
dc.contributor.authorBai, Xiao
dc.contributor.authorLiu, Yun
dc.contributor.authorZhou, Jun
dc.date.accessioned2020-07-10T02:23:42Z
dc.date.available2020-07-10T02:23:42Z
dc.date.issued2019
dc.identifier.issn0167-8655
dc.identifier.doi10.1016/j.patrec.2019.09.021
dc.identifier.urihttp://hdl.handle.net/10072/395308
dc.description.abstractRemarkable performance of the homography estimation has been achieved by the deep CNN based approaches. These homography estimation methods, more often than not, are supervised methods and rely too much on the ground truth annotations as they aim to learn the mapping between image pairs and homography. On the other hand, the inherent invertibility of homography is helpful to avoid over-fitting and improve the performance, which however is ignored by previous homography estimation methods. In this paper, we propose a novel homography estimation approach, named “Self-Supervised Regression Network(SSR-Net)”, which relaxes the need of ground truth annotations and takes advantage of invertibility constraints. We utilize spatial pyramid pooling modules to improve the quality of extracted features in each image by exploiting context information. To employ the invertibility constraints, we adopt the matrix representation of the homography rather than the commonly used 4-point parameterization in other methods. Our proposed SSR-Net produce homography matrices and synthetic images in a cycled way. The network are trained in a self-supervised way by minimizing the combination of photometric loss and invertibility loss. Experiments on the synthetic dataset generated from MSCOCO dataset show that our proposed method outperforms several state-of-the-art approaches.
dc.description.peerreviewedYes
dc.languageEnglish
dc.language.isoeng
dc.publisherElsevier
dc.relation.ispartofpagefrom355
dc.relation.ispartofpageto360
dc.relation.ispartofjournalPattern Recognition Letters
dc.relation.ispartofvolume128
dc.subject.fieldofresearchArtificial intelligence
dc.subject.fieldofresearchCognitive and computational psychology
dc.subject.fieldofresearchcode4602
dc.subject.fieldofresearchcode5204
dc.subject.keywordsScience & Technology
dc.subject.keywordsTechnology
dc.subject.keywordsComputer Science, Artificial Intelligence
dc.subject.keywordsComputer Science
dc.subject.keywordsNomography estimation
dc.titleSelf-Supervised deep homography estimation with invertibility constraints
dc.typeJournal article
dc.type.descriptionC1 - Articles
dcterms.bibliographicCitationWang, C; Wang, X; Bai, X; Liu, Y; Zhou, J, Self-Supervised deep homography estimation with invertibility constraints, Pattern Recognition Letters, 2019, 128, pp. 355-360
dc.date.updated2020-07-10T02:22:38Z
gro.hasfulltextNo Full Text
gro.griffith.authorZhou, Jun


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

  • Journal articles
    Contains articles published by Griffith authors in scholarly journals.

Show simple item record