Rethinking Continual Anomaly Detection on the Edge: Benchmarking Under Realistic Industrial Conditions
For practitioners deploying anomaly detection on edge devices in evolving industrial settings, this work provides a rigorous benchmark and a simple, effective method that challenges the need for complex continual learning techniques.
The paper identifies three critical gaps in continual anomaly detection (CAD) for industrial edge deployment and introduces a unified benchmark with realistic protocols. The proposed DINOSaur method, a training-free approach using a frozen DINOv3 backbone, achieves zero forgetting, outperforms all existing CAD methods across five protocols, and runs under 100 ms inference on an NVIDIA Jetson Orin Nano with on-device adaptation under 30 seconds.
Continual anomaly detection (CAD) addresses the need for industrial inspection systems to adapt to evolving production conditions, yet existing methods share three critical gaps: unrealistic evaluation, no systematic comparison, and no consideration of edge deployment constraints. We introduce a unified benchmark combining discrete-task evaluation on structural and logical anomalies, a novel continuous drift protocol, the first head-to-head comparison of all published CAD methods, and computational efficiency profiling on edge hardware. Our results reveal that existing CAD methods do not consistently outperform traditional approaches with simple experience replay. Thus motivated, we propose DINOSaur, a training-free method combining a frozen DINOv3 backbone with spatially-indexed coreset memory and neighborhood-restricted anomaly scoring. DINOSaur achieves zero forgetting by construction, outperforms all evaluated methods across all five protocols, and runs at sub-100\,ms inference on an NVIDIA Jetson Orin Nano, with on-device adaptation to new tasks in under 30 seconds.