SPG: Sparse-Projected Guides with Sparse Autoencoders for Zero-Shot Anomaly Detection
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915912609693696 |
|---|---|
| author | Nanaumi, Tomoyasu Tsuzuki, Yukino Okubo, Junichi Fujii, Junichiro Yamashita, Takayoshi |
| author_facet | Nanaumi, Tomoyasu Tsuzuki, Yukino Okubo, Junichi Fujii, Junichiro Yamashita, Takayoshi |
| contents | We study zero-shot anomaly detection and segmentation using frozen foundation model features, where all learnable parameters are trained only on a labeled auxiliary dataset and deployed to unseen target categories without any target-domain adaptation. Existing prompt-based approaches use handcrafted or learned prompt embeddings as reference vectors for normal/anomalous states. We propose Sparse-Projected Guides (SPG), a prompt-free framework that learns sparse guide coefficients in the Sparse Autoencoder (SAE) latent space, which generate normal/anomaly guide vectors via the SAE dictionary. SPG employs a two stage learning strategy on the labeled auxiliary dataset: (i) train an SAE on patch-token features, and (ii) optimize only guide coefficients using auxiliary pixel-level masks while freezing the backbone and SAE. On MVTec AD and VisA under cross-dataset zero-shot settings, SPG achieves competitive image-level detection and strong pixel-level segmentation; with DINOv3, SPG attains the highest pixellevel AUROC among the compared methods. We also report SPG instantiated with OpenCLIP (ViT-L/14@336px) to align the backbone with CLIP-based baselines. Moreover, the learned guide coefficients trace decisions back to a small set of dictionary atoms, revealing category-general and category-specific factors. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_02871 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SPG: Sparse-Projected Guides with Sparse Autoencoders for Zero-Shot Anomaly Detection Nanaumi, Tomoyasu Tsuzuki, Yukino Okubo, Junichi Fujii, Junichiro Yamashita, Takayoshi Computer Vision and Pattern Recognition We study zero-shot anomaly detection and segmentation using frozen foundation model features, where all learnable parameters are trained only on a labeled auxiliary dataset and deployed to unseen target categories without any target-domain adaptation. Existing prompt-based approaches use handcrafted or learned prompt embeddings as reference vectors for normal/anomalous states. We propose Sparse-Projected Guides (SPG), a prompt-free framework that learns sparse guide coefficients in the Sparse Autoencoder (SAE) latent space, which generate normal/anomaly guide vectors via the SAE dictionary. SPG employs a two stage learning strategy on the labeled auxiliary dataset: (i) train an SAE on patch-token features, and (ii) optimize only guide coefficients using auxiliary pixel-level masks while freezing the backbone and SAE. On MVTec AD and VisA under cross-dataset zero-shot settings, SPG achieves competitive image-level detection and strong pixel-level segmentation; with DINOv3, SPG attains the highest pixellevel AUROC among the compared methods. We also report SPG instantiated with OpenCLIP (ViT-L/14@336px) to align the backbone with CLIP-based baselines. Moreover, the learned guide coefficients trace decisions back to a small set of dictionary atoms, revealing category-general and category-specific factors. |
| title | SPG: Sparse-Projected Guides with Sparse Autoencoders for Zero-Shot Anomaly Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.02871 |