EUVIMEDEUVIMED entdecken
Crossref

CGR-Net: Consistency Guided ResFormer for Two-View Correspondence Learning

Changcai Yang, Xiaojie Li, Jiayi Ma, Fengyuan Zhuang, Lifang Wei, Riqing Chen, Guodong Chen

IEEE Transactions on Circuits and Systems for Video Technology · 2024-12 · 34 (12) · 12450-12465

DOI: 10.1109/tcsvt.2024.3439348 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2024-12. Verfasst von Changcai Yang und 6 weiteren Autor:innen. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 63 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Fund of Fujian Province, Fujian Innovation Fund Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurately identifying correct correspondences (inliers) in two-view images is a fundamental task in computer vision. Recent studies usually adopt Graph Neural Networks or stack local graphs into global ones to establish neighborhood relations. However, the smoothing properties of Graph Convolutional Neural network (GCN) cause the model to fall into local extreme, which leads to the issue of indistinguishability between inliers and outliers. Especially when the initial correspondences contain a large number of incorrect correspondences (outliers), these studies suffer from severe performance degradation. To address the above issues and refocus perspective information on distinct features, we design a Consistency Guided ResFormer Network (CGR-Net) that uses consistent correspondences to guide model perspective focusing, thereby avoiding the negative impact of outliers. Specifically, we design an efficient Graph Score Calculation module, which aims to compute global graph scores by enhancing the representation of important features and comprehensively capturing the contextual relationships between correspondences. Then, we propose a Consistency Guided Correspondences Selection module to dynamically fuse global graph scores and consistency graphs and construct a novel consistency matrix to accurately recognize inliers. Extensive experiments on various challenging tasks demonstrate that our CGR-Net outperforms state-of-the-art methods. Our code is released at https://github.com/XiaojieLi11/CGR-Net.

Förderung

  • National Natural Science Foundation of China
  • Natural Science Fund of Fujian Province
  • Fujian Innovation Fund Project
  • Fujian Province Science and Technology Plan Guided
  • Big Data in Agroforestry (Cross-Disciplinary) of Fujian Agriculture and Forestry University
  • Science and Technology Innovation Special Fund Project of FAFU

Zitieren

APA

Yang, C., Li, X., Ma, J., Zhuang, F., Wei, L., Chen, R., & Chen, G. (2024). CGR-Net: Consistency Guided ResFormer for Two-View Correspondence Learning. IEEE Transactions on Circuits and Systems for Video Technology, 34(12), 12450-12465. https://doi.org/10.1109/tcsvt.2024.3439348

Vancouver

Yang C, Li X, Ma J, Zhuang F, Wei L, Chen R, et al. CGR-Net: Consistency Guided ResFormer for Two-View Correspondence Learning. IEEE Transactions on Circuits and Systems for Video Technology. 2024;34(12):12450-12465. doi:10.1109/tcsvt.2024.3439348

BibTeX
@article{Yang2024,
  author = {Yang, Changcai and Li, Xiaojie and Ma, Jiayi and Zhuang, Fengyuan and Wei, Lifang and Chen, Riqing and Chen, Guodong},
  title = {CGR-Net: Consistency Guided ResFormer for Two-View Correspondence Learning},
  journal = {IEEE Transactions on Circuits and Systems for Video Technology},
  year = {2024},
  volume = {34},
  number = {12},
  pages = {12450--12465},
  doi = {10.1109/tcsvt.2024.3439348}
}
Kostenlos starten →