AdapTS: Lightweight Teacher-Student Approach for Multi-Class and Continual Visual Anomaly Detection
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911525507170304 |
|---|---|
| author | Barusco, Manuel Pezze, Davide Dalle Borsatti, Francesco Susto, Gian Antonio |
| author_facet | Barusco, Manuel Pezze, Davide Dalle Borsatti, Francesco Susto, Gian Antonio |
| contents | Visual Anomaly Detection (VAD) is crucial for industrial inspection, yet most existing methods are limited to single-category scenarios, failing to address the multi-class and continual learning demands of real-world environments. While Teacher-Student (TS) architectures are efficient, they remain unexplored for the Continual Setting. To bridge this gap, we propose AdapTS, a unified TS framework designed for multi-class and continual settings, optimized for edge deployment. AdapTS eliminates the need for two different architectures by utilizing a single shared frozen backbone and injecting lightweight trainable adapters into the student pathway. Training is enhanced via a segmentation-guided objective and synthetic Perlin noise, while a prototype-based task identification mechanism dynamically selects adapters at inference with 99\% accuracy.
Experiments on MVTec AD and VisA demonstrate that AdapTS matches the performance of existing TS methods across multi-class and continual learning scenarios, while drastically reducing memory overhead. Our lightest variant, AdapTS-S, requires only 8 MB of additional memory, 13x less than STFPM (95 MB), 48x less than RD4AD (360 MB), and 149x less than DeSTSeg (1120 MB), making it a highly scalable solution for edge deployment in complex industrial environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_17530 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | AdapTS: Lightweight Teacher-Student Approach for Multi-Class and Continual Visual Anomaly Detection Barusco, Manuel Pezze, Davide Dalle Borsatti, Francesco Susto, Gian Antonio Computer Vision and Pattern Recognition Artificial Intelligence Visual Anomaly Detection (VAD) is crucial for industrial inspection, yet most existing methods are limited to single-category scenarios, failing to address the multi-class and continual learning demands of real-world environments. While Teacher-Student (TS) architectures are efficient, they remain unexplored for the Continual Setting. To bridge this gap, we propose AdapTS, a unified TS framework designed for multi-class and continual settings, optimized for edge deployment. AdapTS eliminates the need for two different architectures by utilizing a single shared frozen backbone and injecting lightweight trainable adapters into the student pathway. Training is enhanced via a segmentation-guided objective and synthetic Perlin noise, while a prototype-based task identification mechanism dynamically selects adapters at inference with 99\% accuracy. Experiments on MVTec AD and VisA demonstrate that AdapTS matches the performance of existing TS methods across multi-class and continual learning scenarios, while drastically reducing memory overhead. Our lightest variant, AdapTS-S, requires only 8 MB of additional memory, 13x less than STFPM (95 MB), 48x less than RD4AD (360 MB), and 149x less than DeSTSeg (1120 MB), making it a highly scalable solution for edge deployment in complex industrial environments. |
| title | AdapTS: Lightweight Teacher-Student Approach for Multi-Class and Continual Visual Anomaly Detection |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2603.17530 |