Large-Scale Universal Defect Generation: Foundation Models and Datasets
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917398475440128 |
|---|---|
| author | Fan, Yuanting Liu, Jun Gao, Bin-Bin Chen, Xiaochen Lin, Yuhuan Dai, Zhewei Zhan, Jiawei Wang, Chengjie |
| author_facet | Fan, Yuanting Liu, Jun Gao, Bin-Bin Chen, Xiaochen Lin, Yuhuan Dai, Zhewei Zhan, Jiawei Wang, Chengjie |
| contents | Existing defect/anomaly generation methods often rely on few-shot learning, which overfits to specific defect categories due to the lack of large-scale paired defect editing data. This issue is aggravated by substantial variations in defect scale and morphology, resulting in limited generalization, degraded realism, and category consistency. We address these challenges by introducing UDG, a large-scale dataset of 300K normal-abnormal-mask-caption quadruplets spanning diverse domains, and by presenting UniDG, a universal defect generation foundation model that supports both reference-based defect generation and text instruction-based defect editing without per-category fine-tuning. UniDG performs Defect-Context Editing via adaptive defect cropping and structured diptych input format, and fuses reference and target conditions through MM-DiT multimodal attention. A two-stage training strategy, Diversity-SFT followed by Consistency-RFT, further improves diversity while enhancing realism and reference consistency. Extensive experiments on MVTec-AD and VisA show that UniDG outperforms prior few-shot anomaly generation and image insertion/editing baselines in synthesis quality and downstream single- and multi-class anomaly detection/localization. Code will be available at https://github.com/RetoFan233/UniDG. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_08915 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Large-Scale Universal Defect Generation: Foundation Models and Datasets Fan, Yuanting Liu, Jun Gao, Bin-Bin Chen, Xiaochen Lin, Yuhuan Dai, Zhewei Zhan, Jiawei Wang, Chengjie Computer Vision and Pattern Recognition Artificial Intelligence Existing defect/anomaly generation methods often rely on few-shot learning, which overfits to specific defect categories due to the lack of large-scale paired defect editing data. This issue is aggravated by substantial variations in defect scale and morphology, resulting in limited generalization, degraded realism, and category consistency. We address these challenges by introducing UDG, a large-scale dataset of 300K normal-abnormal-mask-caption quadruplets spanning diverse domains, and by presenting UniDG, a universal defect generation foundation model that supports both reference-based defect generation and text instruction-based defect editing without per-category fine-tuning. UniDG performs Defect-Context Editing via adaptive defect cropping and structured diptych input format, and fuses reference and target conditions through MM-DiT multimodal attention. A two-stage training strategy, Diversity-SFT followed by Consistency-RFT, further improves diversity while enhancing realism and reference consistency. Extensive experiments on MVTec-AD and VisA show that UniDG outperforms prior few-shot anomaly generation and image insertion/editing baselines in synthesis quality and downstream single- and multi-class anomaly detection/localization. Code will be available at https://github.com/RetoFan233/UniDG. |
| title | Large-Scale Universal Defect Generation: Foundation Models and Datasets |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2604.08915 |