DC Field | Value | Language |
---|---|---|
dc.contributor.author | JONGMIN, LEE | - |
dc.contributor.author | KIM, BYUNGJIN | - |
dc.contributor.author | KIM, SEUNG WOOK | - |
dc.contributor.author | CHO, MINSU | - |
dc.date.accessioned | 2024-03-05T09:12:11Z | - |
dc.date.available | 2024-03-05T09:12:11Z | - |
dc.date.created | 2024-03-04 | - |
dc.date.issued | 2023-06 | - |
dc.identifier.uri | https://oasis.postech.ac.kr/handle/2014.oak/121034 | - |
dc.description.abstract | Extracting discriminative local features that are invariant to imaging variations is an integral part of establishing correspondences between images. In this work, we introduce a self-supervised learning framework to extract discriminative rotation-invariant descriptors using group-equivariant CNNs. Thanks to employing group-equivariant CNNs, our method effectively learns to obtain rotation-equivariant features and their orientations explicitly, without having to perform sophisticated data augmentations. The resultant features and their orientations are further processed by group aligning, a novel invariant mapping technique that shifts the group-equivariant features by their orientations along the group dimension. Our group aligning technique achieves rotation-invariance without any collapse of the group dimension and thus eschews loss of discriminability. The proposed method is trained end-to-end in a self-supervised manner, where we use an orientation alignment loss for the orientation estimation and a contrastive descriptor loss for robust local descriptors to geometric/photometric variations. Our method demonstrates state-of-the-art matching accuracy among existing rotation-invariant descriptors under varying rotation and also shows competitive results when transferred to the task of keypoint matching and camera pose estimation. | - |
dc.language | English | - |
dc.publisher | IEEE Computer Society | - |
dc.relation.isPartOf | 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023 | - |
dc.relation.isPartOf | Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition | - |
dc.title | Learning Rotation-Equivariant Features for Visual Correspondence | - |
dc.type | Conference | - |
dc.type.rims | CONF | - |
dc.identifier.bibliographicCitation | 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023, pp.21887 - 21897 | - |
dc.citation.conferenceDate | 2023-06-18 | - |
dc.citation.conferencePlace | CA | - |
dc.citation.endPage | 21897 | - |
dc.citation.startPage | 21887 | - |
dc.citation.title | 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023 | - |
dc.contributor.affiliatedAuthor | JONGMIN, LEE | - |
dc.contributor.affiliatedAuthor | KIM, BYUNGJIN | - |
dc.contributor.affiliatedAuthor | KIM, SEUNG WOOK | - |
dc.contributor.affiliatedAuthor | CHO, MINSU | - |
dc.description.journalClass | 1 | - |
dc.description.journalClass | 1 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
library@postech.ac.kr Tel: 054-279-2548
Copyrights © by 2017 Pohang University of Science ad Technology All right reserved.