MECAD: A multi-expert architecture for continual anomaly detection

2025/12/17 by Dahmardeh, Malihe, Setti, Francesco
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2512.15323

Abstract

In this paper we propose MECAD, a novel approach for continual anomaly detection using a multi-expert architecture. Our system dynamically assigns experts to object classes based on feature similarity and employs efficient memory management to preserve the knowledge of previously seen classes. By leveraging an optimized coreset selection and a specialized replay buffer mechanism, we enable incremental learning without requiring full model retraining. Our experimental evaluation on the MVTec AD dataset demonstrates that the optimal 5-expert configuration achieves an average AUROC of 0.8259 across 15 diverse object categories while significantly reducing knowledge degradation compared to single-expert approaches. This framework balances computational efficiency, specialized knowledge retention, and adaptability, making it well-suited for industrial environments with evolving product types.

Citations

• Theory on Mixture-of-Experts in Continual Learning
• Mixture of Experts Meets Prompt-Based Continual Learning
• Unveiling the Anomalies in an Ever-Changing World: A Benchmark for Pixel-Level Anomaly Detection in Continual Learning
• Continuous Memory Representation for Anomaly Detection
• Divide and not forget: Ensemble of selectively trained experts in Continual Learning
• Unsupervised Continual Anomaly Detection with Contrastively-learned Prompt
• An Incremental Unified Framework for Small Defect Inspection
• Continual Learning Approaches for Anomaly Detection
• TAME: Task Agnostic Continual Learning using Multiple Experts
• ADMoE: Anomaly Detection with Mixture-of-Experts from Noisy Labels
• Mixture-of-Variational-Experts for Continual Learning
• Towards Total Recall in Industrial Anomaly Detection
• Online Coreset Selection for Rehearsal-based Continual Learning
• PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and\n Localization
• Gradient based sample selection for online continual learning
• Experience Replay for Continual Learning
• Memory Aware Synapses: Learning what (not) to forget
• Variational Continual Learning
• iCaRL: Incremental Classifier and Representation Learning
• Expert Gate: Lifelong Learning with a Network of Experts

Related

• CADIC: Continual Anomaly Detection Based on Incremental Coreset
• C3D-AD: Toward Continual 3D Anomaly Detection via Kernel Attention with Learnable Advisor
• Unsupervised Continual Anomaly Detection with Contrastively-learned Prompt
• Pseudo Replay-based Class Continual Learning for Online New Category Anomaly Detection in Advanced Manufacturing
• Towards Principled Continual Anomaly Detection: A Systematic Framework and Benchmark Scenarios