CRFT: Consistent-Recurrent Feature Flow Transformer for Cross-Modal Image Registration
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908942477557760 |
|---|---|
| author | Liu, Xuecong Ding, Mengzhu Sun, Zixuan Li, Zhang Teng, Xichao |
| author_facet | Liu, Xuecong Ding, Mengzhu Sun, Zixuan Li, Zhang Teng, Xichao |
| contents | We present Consistent-Recurrent Feature Flow Transformer (CRFT), a unified coarse-to-fine framework based on feature flow learning for robust cross-modal image registration. CRFT learns a modality-independent feature flow representation within a transformer-based architecture that jointly performs feature alignment and flow estimation. The coarse stage establishes global correspondences through multi-scale feature correlation, while the fine stage refines local details via hierarchical feature fusion and adaptive spatial reasoning. To enhance geometric adaptability, an iterative discrepancy-guided attention mechanism with a Spatial Geometric Transform (SGT) recurrently refines the flow field, progressively capturing subtle spatial inconsistencies and enforcing feature-level consistency. This design enables accurate alignment under large affine and scale variations while maintaining structural coherence across modalities. Extensive experiments on diverse cross-modal datasets demonstrate that CRFT consistently outperforms state-of-the-art registration methods in both accuracy and robustness. Beyond registration, CRFT provides a generalizable paradigm for multimodal spatial correspondence, offering broad applicability to remote sensing, autonomous navigation, and medical imaging. Code and datasets are publicly available at https://github.com/NEU-Liuxuecong/CRFT. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_05689 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | CRFT: Consistent-Recurrent Feature Flow Transformer for Cross-Modal Image Registration Liu, Xuecong Ding, Mengzhu Sun, Zixuan Li, Zhang Teng, Xichao Computer Vision and Pattern Recognition Artificial Intelligence We present Consistent-Recurrent Feature Flow Transformer (CRFT), a unified coarse-to-fine framework based on feature flow learning for robust cross-modal image registration. CRFT learns a modality-independent feature flow representation within a transformer-based architecture that jointly performs feature alignment and flow estimation. The coarse stage establishes global correspondences through multi-scale feature correlation, while the fine stage refines local details via hierarchical feature fusion and adaptive spatial reasoning. To enhance geometric adaptability, an iterative discrepancy-guided attention mechanism with a Spatial Geometric Transform (SGT) recurrently refines the flow field, progressively capturing subtle spatial inconsistencies and enforcing feature-level consistency. This design enables accurate alignment under large affine and scale variations while maintaining structural coherence across modalities. Extensive experiments on diverse cross-modal datasets demonstrate that CRFT consistently outperforms state-of-the-art registration methods in both accuracy and robustness. Beyond registration, CRFT provides a generalizable paradigm for multimodal spatial correspondence, offering broad applicability to remote sensing, autonomous navigation, and medical imaging. Code and datasets are publicly available at https://github.com/NEU-Liuxuecong/CRFT. |
| title | CRFT: Consistent-Recurrent Feature Flow Transformer for Cross-Modal Image Registration |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2604.05689 |