Title: FAF-CD: Frequency-Aware Fusion for Change Detection under Imperfect Multimodal Remote Sensing

Original: arXiv:2606.03114v1 Announce Type: new Abstract: Remote sensing change detection for real-world monitoring often relies on imperfect heterogeneous observations, where pre- and post-event images may be asynchronous, cross-sensor, or affected by illumination, seasonal, and modality shifts. This setting is especially challenging for EO-SAR disaster mapping, where nuisance variation can resemble structural damage. We propose FAF-CD, a frequency-aware hybrid framework with a DINOv3-pretrained ConvNeXt encoder and a linear-complexity VMamba-based decoder. Its rectification-aware tri-branch fusion module combines deformable spatial alignment with Fourier and Haar-wavelet comparisons, using adaptive gating to aggregate complementary cues across scales. On BRIGHT validation, a matched heterogeneous EO-SAR adaptation improves clean and perturbed tc-mIoU/tc-mAP over NeXt2Former-CD. FAF-CD also generalizes to binary optical CD, achieving 0.924 cF1 on LEVIR-CD and 0.955 cF1 on WHU-CD, and obtains the best average perturbed cIoU/cF1 on both binary datasets among M-CD and NeXt2Former-CD under pseudo-change-aligned stress tests. It further reduces cost by approximately 24 GFLOPs relative to NeXt2Former-CD while maintaining or improving accuracy.

Rewrite: Title: FAF-CD: Frequency-Aware Fusion for Change Detection under Imperfect Multimodal Remote Sensing

Original: arXiv:2606.03114v1 Announce Type: new Abstract: In practical remote sensing applications, change detection frequently encounters imperfect, heterogeneous data. Pre- and post-event imagery may suffer from asynchronicity, cross-sensor discrepancies, or variations in illumination, seasonality, and modality. These conditions pose significant difficulties for EO-SAR disaster mapping, where environmental noise can mimic actual structural damage. To address these challenges, we introduce FAF-CD, a hybrid framework that integrates a DINOv3-pretrained ConvNeXt encoder with a linear-complexity VMamba decoder. Central to this approach is a rectification-aware tri-branch fusion module, which merges deformable spatial alignment with Fourier and Haar-wavelet analyses. Adaptive gating mechanisms are employed to synthesize complementary information across multiple scales. Evaluation on the BRIGHT dataset demonstrates that our heterogeneous EO-SAR adaptation surpasses NeXt2Former-CD in both clean and perturbed tc-mIoU and tc-mAP metrics. Furthermore, FAF-CD exhibits strong generalization to binary optical change detection, securing cF1 scores of 0.924 on LEVIR-CD and 0.955 on WHU-CD. Under pseudo-change-aligned stress tests, it achieves the highest average perturbed cIoU and cF1 among M-CD and NeXt2Former-CD on both binary benchmarks. Notably, FAF-CD cuts computational cost by roughly 24 GFLOPs compared to NeXt2Former-CD, all while preserving or enhancing predictive accuracy.

Source: arXiv Generated at: 2026-06-03 00:00:00 UTC