Learning to Identify Correct 2D-2D Line Correspondences on Sphere
- Authors
- Li, Haoang; Chen, Kai; Zhao, Ji; Wang, Jiangliu; Kim, Pyojin; Liu, Zhe; Liu, Yun-Hui
- Issue Date
- Jun-2021
- Publisher
- IEEE COMPUTER SOC
- Citation
- 2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, pp 11738 - 11747
- Pages
- 10
- Journal Title
- 2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
- Start Page
- 11738
- End Page
- 11747
- URI
- https://scholarworks.sookmyung.ac.kr/handle/2020.sw.sookmyung/146592
- DOI
- 10.1109/CVPR46437.2021.01157
- ISSN
- 1063-6919
- Abstract
- Given a set of putative 2D-2D line correspondences, we aim to identify correct matches. Existing methods exploit the geometric constraints. They are only applicable to structured scenes with orthogonality, parallelism and coplanarity. In contrast, we propose the first approach suitable for both structured and unstructured scenes. Instead of geometric constraint, we leverage the spatial regularity on sphere. Specifically, we propose to map line correspondences into vectors tangent to sphere. We use these vectors to encode both angular and positional variations of image lines, which is more reliable and concise than directly using inclinations, midpoints or endpoints of image lines. Neighboring vectors mapped from correct matches exhibit a spatial regularity called local trend consistency, regardless of the type of scenes. To encode this regularity, we design a neural network and also propose a novel loss function that enforces the smoothness constraint of vector field. In addition, we establish a large real-world dataset for image line matching. Experiments showed that our approach outperforms state-of-the-art ones in terms of accuracy, efficiency and robustness, and also leads to high generalization.
- Files in This Item
-
- Appears in
Collections - 공과대학 > 기계시스템학부 > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.