SuperAD: A Training-free Anomaly Classification and Segmentation Method for CVPR 2025 VAND 3.0 Workshop Challenge Track 1: Adapt & Detect
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910970310295552 |
|---|---|
| author | Zhang, Huaiyuan Chen, Hang Cheng, Yu Wu, Shunyi Sun, Linghao Han, Linao Shi, Zeyu Qi, Lei |
| author_facet | Zhang, Huaiyuan Chen, Hang Cheng, Yu Wu, Shunyi Sun, Linghao Han, Linao Shi, Zeyu Qi, Lei |
| contents | In this technical report, we present our solution to the CVPR 2025 Visual Anomaly and Novelty Detection (VAND) 3.0 Workshop Challenge Track 1: Adapt & Detect: Robust Anomaly Detection in Real-World Applications. In real-world industrial anomaly detection, it is crucial to accurately identify anomalies with physical complexity, such as transparent or reflective surfaces, occlusions, and low-contrast contaminations. The recently proposed MVTec AD 2 dataset significantly narrows the gap between publicly available benchmarks and anomalies found in real-world industrial environments. To address the challenges posed by this dataset--such as complex and varying lighting conditions and real anomalies with large scale differences--we propose a fully training-free anomaly detection and segmentation method based on feature extraction using the DINOv2 model named SuperAD. Our method carefully selects a small number of normal reference images and constructs a memory bank by leveraging the strong representational power of DINOv2. Anomalies are then segmented by performing nearest neighbor matching between test image features and the memory bank. Our method achieves competitive results on both test sets of the MVTec AD 2 dataset. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_19750 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SuperAD: A Training-free Anomaly Classification and Segmentation Method for CVPR 2025 VAND 3.0 Workshop Challenge Track 1: Adapt & Detect Zhang, Huaiyuan Chen, Hang Cheng, Yu Wu, Shunyi Sun, Linghao Han, Linao Shi, Zeyu Qi, Lei Computer Vision and Pattern Recognition In this technical report, we present our solution to the CVPR 2025 Visual Anomaly and Novelty Detection (VAND) 3.0 Workshop Challenge Track 1: Adapt & Detect: Robust Anomaly Detection in Real-World Applications. In real-world industrial anomaly detection, it is crucial to accurately identify anomalies with physical complexity, such as transparent or reflective surfaces, occlusions, and low-contrast contaminations. The recently proposed MVTec AD 2 dataset significantly narrows the gap between publicly available benchmarks and anomalies found in real-world industrial environments. To address the challenges posed by this dataset--such as complex and varying lighting conditions and real anomalies with large scale differences--we propose a fully training-free anomaly detection and segmentation method based on feature extraction using the DINOv2 model named SuperAD. Our method carefully selects a small number of normal reference images and constructs a memory bank by leveraging the strong representational power of DINOv2. Anomalies are then segmented by performing nearest neighbor matching between test image features and the memory bank. Our method achieves competitive results on both test sets of the MVTec AD 2 dataset. |
| title | SuperAD: A Training-free Anomaly Classification and Segmentation Method for CVPR 2025 VAND 3.0 Workshop Challenge Track 1: Adapt & Detect |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.19750 |