UniADC: A Unified Framework for Anomaly Detection and Classification

2025-11-10

AI Summary

Existing approaches treat anomaly detection and fine-grained classification as disjoint tasks, ignoring their intrinsic correlations—leading to suboptimal performance in few-shot and zero-shot settings. This paper proposes the first unified framework for joint anomaly detection and fine-grained classification. We design a training-free, controllable image inpainting network that synthesizes class-conditional anomalous images without requiring any real anomaly samples, enabling effective data augmentation and cross-task knowledge sharing. Furthermore, we introduce a multi-task discriminator that jointly optimizes anomaly localization and category recognition, enhanced by an anomaly-prior-guided region inpainting mechanism and fine-grained feature alignment. Extensive experiments on three benchmarks—MVTec-FS, MTD, and WFDD—demonstrate significant improvements over state-of-the-art methods, achieving superior integrated performance in anomaly detection, precise localization, and fine-grained classification.

Technology Category

• Computer Vision: Object Detection & Categorization
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

In this paper, we introduce the task of unified anomaly detection and classification, which aims to simultaneously detect anomalous regions in images and identify their specific categories. Existing methods typically treat anomaly detection and classification as separate tasks, thereby neglecting their inherent correlation, limiting information sharing, and resulting in suboptimal performance. To address this, we propose UniADC, a unified anomaly detection and classification model that can effectively perform both tasks with only a few or even no anomaly images. Specifically, UniADC consists of two key components: a training-free controllable inpainting network and a multi-task discriminator. The inpainting network can synthesize anomaly images of specific categories by repainting normal regions guided by anomaly priors, and can also repaint few-shot anomaly samples to augment the available anomaly data. The multi-task discriminator is then trained on these synthesized samples, enabling precise anomaly detection and classification by aligning fine-grained image features with anomaly-category embeddings. We conduct extensive experiments on three anomaly detection and classification datasets, including MVTec-FS, MTD, and WFDD, and the results demonstrate that UniADC consistently outperforms existing methods in anomaly detection, localization, and classification. The code is available at https://github.com/cnulab/UniADC.

Problem

• Unified framework for simultaneous anomaly detection and classification in images
• Addresses limitations of separate anomaly detection and classification approaches
• Performs both tasks effectively with few or no anomaly images

Innovation

• UniADC combines anomaly detection and classification tasks
• Uses controllable inpainting network to synthesize anomaly images
• Employs multi-task discriminator with category embeddings alignment