ABounD: Adversarial Boundary-Driven Few-Shot Learning for Multi-Class Anomaly Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Deng, Runzhi, Hu, Yundi, Zhang, Xinshuang, Wang, Zhao, Liu, Xixi, Dai, Wang-Zhou, Shan, Caifeng, Zhao, Fang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912965408587776
author Deng, Runzhi
Hu, Yundi
Zhang, Xinshuang
Wang, Zhao
Liu, Xixi
Dai, Wang-Zhou
Shan, Caifeng
Zhao, Fang
author_facet Deng, Runzhi
Hu, Yundi
Zhang, Xinshuang
Wang, Zhao
Liu, Xixi
Dai, Wang-Zhou
Shan, Caifeng
Zhao, Fang
contents Few-shot multi-class industrial anomaly detection identifies diverse defects across multiple categories using a single unified model and limited normal samples. Although vision-language models offer strong generalization, modeling multiple distinct category manifolds concurrently without actual anomalous data causes feature space collapse and cross-class interference. Consequently, existing methods often fail to balance scalability and precision, leading to either isolated single-class retraining or excessively loose decision margins. To address this limitation, we present a one-for-all learning framework called ABounD that unites semantic concept anchoring with geometric boundary optimization. This method employs two lightweight mechanisms to resolve multi-class ambiguity. First, the Dynamic Concept Fusion module generates class-adaptive semantic anchors via query-aware hierarchical calibration, disentangling overlapping category concepts. Second, using these anchors, the Adversarial Boundary Forging module constructs a tight, class-tailored decision margin by synthesizing adversarial boundary-level fence features to prevent cross-class boundary blurring. Optimized in a single stage, ABounD removes the requirement for isolated per-category retraining in few-shot settings. Experiments on seven industrial benchmarks show that the proposed method achieves state-of-the-art detection and localization performance for multi-class few-shot anomaly detection while maintaining low computational costs during training and inference.
format Preprint
id arxiv_https___arxiv_org_abs_2511_22436
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ABounD: Adversarial Boundary-Driven Few-Shot Learning for Multi-Class Anomaly Detection
Deng, Runzhi
Hu, Yundi
Zhang, Xinshuang
Wang, Zhao
Liu, Xixi
Dai, Wang-Zhou
Shan, Caifeng
Zhao, Fang
Computer Vision and Pattern Recognition
Few-shot multi-class industrial anomaly detection identifies diverse defects across multiple categories using a single unified model and limited normal samples. Although vision-language models offer strong generalization, modeling multiple distinct category manifolds concurrently without actual anomalous data causes feature space collapse and cross-class interference. Consequently, existing methods often fail to balance scalability and precision, leading to either isolated single-class retraining or excessively loose decision margins. To address this limitation, we present a one-for-all learning framework called ABounD that unites semantic concept anchoring with geometric boundary optimization. This method employs two lightweight mechanisms to resolve multi-class ambiguity. First, the Dynamic Concept Fusion module generates class-adaptive semantic anchors via query-aware hierarchical calibration, disentangling overlapping category concepts. Second, using these anchors, the Adversarial Boundary Forging module constructs a tight, class-tailored decision margin by synthesizing adversarial boundary-level fence features to prevent cross-class boundary blurring. Optimized in a single stage, ABounD removes the requirement for isolated per-category retraining in few-shot settings. Experiments on seven industrial benchmarks show that the proposed method achieves state-of-the-art detection and localization performance for multi-class few-shot anomaly detection while maintaining low computational costs during training and inference.
title ABounD: Adversarial Boundary-Driven Few-Shot Learning for Multi-Class Anomaly Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.22436