University of Oulu

K. Li, L. Wang, L. Liu, Q. Ran, K. Xu and Y. Guo, "Decoupling Makes Weakly Supervised Local Feature Better," 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA, 2022, pp. 15817-15827, doi: 10.1109/CVPR52688.2022.01538

Decoupling makes weakly supervised local feature better

Saved in:
Author: Li, Kunhong1,2; Wang, Longguang3; Liu, Li3,4;
Organizations: 1Sun Yat-Sen University
2The Shenzhen Campus of Sun Yat-Sen University
3National University of Defense Technology
4University of Oulu
5Alibaba Group
Format: article
Version: accepted version
Access: open
Online Access: PDF Full Text (PDF, 7.3 MB)
Persistent link:
Language: English
Published: Institute of Electrical and Electronics Engineers, 2022
Publish Date: 2023-04-11


Weakly supervised learning can help local feature methods to overcome the obstacle of acquiring a large-scale dataset with densely labeled correspondences. However, since weak supervision cannot distinguish the losses caused by the detection and description steps, directly conducting weakly supervised learning within a joint training describe-then-detect pipeline suffers limited performance. In this paper, we propose a decoupled training describe-then-detect pipeline tailored for weakly supervised local feature learning. Within our pipeline, the detection step is decoupled from the description step and postponed until discriminative and robust descriptors are learned. In addition, we introduce a line-to-window search strategy to explicitly use the camera pose information for better descriptor learning. Extensive experiments show that our method, namely PoSFeat (Camera Pose Supervised Feature), outperforms previous fully and weakly supervised methods and achieves state-of-the-art performance on a wide range of downstream task.

see all

Series: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
ISSN: 1063-6919
ISSN-E: 2575-7075
ISSN-L: 1063-6919
ISBN: 978-1-6654-6946-3
ISBN Print: 978-1-6654-6947-0
Pages: 15817 - 15827
DOI: 10.1109/cvpr52688.2022.01538
Host publication: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Conference: IEEE/CVF Conference on Computer Vision and Pattern Recognition
Type of Publication: A4 Article in conference proceedings
Field of Science: 113 Computer and information sciences
Funding: This work was partially supported by the National Key Research and Development Program of China (No. 2021YFB3100800), the Shenzhen Science and Technology Program (No. RCYX20200714114641140), and National Natural Science Foundation of China (No. U20A20185, 61972435, 62132021).
Copyright information: © 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.