CCTformer: Calibrated Context-Aware Transformer for Correspondence Pruning
IEEE Signal Processing Letters · 2026 · 33 · 1461-1465
DOI: 10.1109/lsp.2026.3679280 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Signal Processing Letters, 2026. Verfasst von Weiwei Zhou und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 29 Literaturangaben. Gefördert von: China Postdoctoral Science Foundation, Engineering Research Center of Big Data Intelligence, Ministry of Education, Fuzhou University.
Abstract (Originalsprache)
Correspondence pruning is a core problem in computer vision, aiming to distinguish correct matches from a large set of initial correspondences. Recent Transformer-based methods, such as VSFormer, have shown remarkable performance by jointly modeling visual appearance and spatial relationships. Nevertheless, these methods often depend on abstract, scene-level visual representations, which are insufficiently precise for validating individual correspondences. In addition, their context aggregation modules often struggle to capture multi-granularity corresponding features, particularly under complex geometric variations. To overcome these limitations, we propose the Calibrated Context-aware Transformer with two key components. The Calibrated Visual Cues Extractor samples features directly at putative match locations to provide well-aligned, correspondence-specific visual cues. The Dual-Perspective Graph Transformer replaces conventional graph attention to better model both node-level and edge-level interactions. Together, these modules enable more robust local-global context fusion. Extensive experiments on YFCC100 M and SUN3D show that CCTformer consistently outperforms the baseline and other state-of-the-art methods.
Förderung
- China Postdoctoral Science Foundation
- Engineering Research Center of Big Data Intelligence, Ministry of Education
- Fuzhou University
Zitieren
APA
Zhou, W., Zhao, S., Liu, Y., & Li, Y. (2026). CCTformer: Calibrated Context-Aware Transformer for Correspondence Pruning. IEEE Signal Processing Letters, 33, 1461-1465. https://doi.org/10.1109/lsp.2026.3679280
Vancouver
Zhou W, Zhao S, Liu Y, Li Y. CCTformer: Calibrated Context-Aware Transformer for Correspondence Pruning. IEEE Signal Processing Letters. 2026;33:1461-1465. doi:10.1109/lsp.2026.3679280
BibTeX
@article{Zhou2026,
author = {Zhou, Weiwei and Zhao, Shengjie and Liu, Yizhang and Li, Yanping},
title = {CCTformer: Calibrated Context-Aware Transformer for Correspondence Pruning},
journal = {IEEE Signal Processing Letters},
year = {2026},
volume = {33},
pages = {1461--1465},
doi = {10.1109/lsp.2026.3679280}
}
EUVIMED