CrackSegFlow: Controllable Flow Matching Synthesis for Generalizable Crack Segmentation with a 50K Image-Mask Benchmark
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_ | 1866912817097998336 |
|---|---|
| author | Asadi, Babak Wu, Peiyang Golparvar-Fard, Mani Hajj, Ramez |
| author_facet | Asadi, Babak Wu, Peiyang Golparvar-Fard, Mani Hajj, Ramez |
| contents | Defect segmentation is central to computer vision based inspection of infrastructure assets during both construction and operation. However, deployment remains limited due to scarce pixel-level labels and domain shift across environments. We introduce CrackSegFlow, a controllable Flow Matching synthesis method that renders synthetic images of cracks from masks with pixel-level alignment. Our renderer combines topology-preserving mask injection with edge gating to maintain thin-structure continuity. Class-conditional FM samples masks for topology diversity, and CrackSegFlow renders aligned ground truth images from them. We further inject cracks onto crack-free backgrounds to diversify confounders and reduce false positives. Across five datasets and using a CNN-Transformer backbone, our results demonstrate that adding synthesized pairs improves in-domain performance by +5.37 mIoU and +5.13 F1, while target-guided cross-domain synthesis driven by target mask statistics adds +13.12 mIoU and +14.82 F1. We also release CSF-50K, a benchmark dataset comprising 50,000 image-mask pairs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_03637 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | CrackSegFlow: Controllable Flow Matching Synthesis for Generalizable Crack Segmentation with a 50K Image-Mask Benchmark Asadi, Babak Wu, Peiyang Golparvar-Fard, Mani Hajj, Ramez Computer Vision and Pattern Recognition Defect segmentation is central to computer vision based inspection of infrastructure assets during both construction and operation. However, deployment remains limited due to scarce pixel-level labels and domain shift across environments. We introduce CrackSegFlow, a controllable Flow Matching synthesis method that renders synthetic images of cracks from masks with pixel-level alignment. Our renderer combines topology-preserving mask injection with edge gating to maintain thin-structure continuity. Class-conditional FM samples masks for topology diversity, and CrackSegFlow renders aligned ground truth images from them. We further inject cracks onto crack-free backgrounds to diversify confounders and reduce false positives. Across five datasets and using a CNN-Transformer backbone, our results demonstrate that adding synthesized pairs improves in-domain performance by +5.37 mIoU and +5.13 F1, while target-guided cross-domain synthesis driven by target mask statistics adds +13.12 mIoU and +14.82 F1. We also release CSF-50K, a benchmark dataset comprising 50,000 image-mask pairs. |
| title | CrackSegFlow: Controllable Flow Matching Synthesis for Generalizable Crack Segmentation with a 50K Image-Mask Benchmark |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.03637 |