CCAD: Compressed Global Feature Conditioned Anomaly Detection

2025-12-24

AI Summary

To address poor domain-shift robustness and inefficient training in few-shot industrial anomaly detection, this paper proposes a novel paradigm integrating reconstruction with unsupervised representation learning: pixel-level reconstruction is conditioned on adaptively compressed global semantic features, enabling joint feature disentanglement and strong reconstruction constraints. Key contributions include: (1) the first architecture to use globally compressed semantic features as explicit conditioning signals for reconstruction; (2) a learnable adaptive compression mechanism that enhances generalization while accelerating convergence; and (3) a restructured and re-annotated version of the DAGM 2007 dataset. Experiments demonstrate that our method achieves significant AUC improvements over state-of-the-art approaches across multiple benchmarks, with over 30% faster training convergence. The source code is publicly available.

Technology Category

• Machine Learning: Feature Construction/Reformulation
• Computer Vision: Adversarial Attacks & Robustness
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• 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

Anomaly detection holds considerable industrial significance, especially in scenarios with limited anomalous data. Currently, reconstruction-based and unsupervised representation-based approaches are the primary focus. However, unsupervised representation-based methods struggle to extract robust features under domain shift, whereas reconstruction-based methods often suffer from low training efficiency and performance degradation due to insufficient constraints. To address these challenges, we propose a novel method named Compressed Global Feature Conditioned Anomaly Detection (CCAD). CCAD synergizes the strengths of both paradigms by adapting global features as a new modality condition for the reconstruction model. Furthermore, we design an adaptive compression mechanism to enhance both generalization and training efficiency. Extensive experiments demonstrate that CCAD consistently outperforms state-of-the-art methods in terms of AUC while achieving faster convergence. In addition, we contribute a reorganized and re-annotated version of the DAGM 2007 dataset with new annotations to further validate our method's effectiveness. The code for reproducing main results is available at https://github.com/chloeqxq/CCAD.

Problem

• Addresses anomaly detection with limited anomalous data
• Improves feature robustness and training efficiency
• Outperforms state-of-the-art methods in AUC and convergence

Innovation

• Combines global features as reconstruction condition
• Uses adaptive compression for generalization and efficiency
• Outperforms state-of-the-art methods in AUC and convergence