Continual Visual Anomaly Detection on the Edge: Benchmark and Efficient Solutions

TL;DR

This paper introduces a comprehensive benchmark for visual anomaly detection on edge devices in continual learning scenarios, and proposes Tiny-Dinomaly, a lightweight, efficient model that outperforms existing methods in resource-constrained settings.

Contribution

It provides the first benchmark for edge-based continual VAD and introduces Tiny-Dinomaly, a novel lightweight model with improved efficiency and accuracy.

Findings

Tiny-Dinomaly achieves 13x smaller memory footprint and 20x lower computational cost.

The benchmark evaluates seven VAD models across three lightweight backbones.

Targeted modifications improve PatchCore and PaDiM efficiency in continual learning.

Abstract

Visual Anomaly Detection (VAD) is a critical task for many applications including industrial inspection and healthcare. While VAD has been extensively studied, two key challenges remain largely unaddressed in conjunction: edge deployment, where computational resources are severely constrained, and continual learning, where models must adapt to evolving data distributions without forgetting previously acquired knowledge. Our benchmark provides guidance for the selection of the optimal backbone and VAD method under joint efficiency and adaptability constraints, characterizing the trade-offs between memory footprint, inference cost, and detection performance. Studying these challenges in isolation is insufficient, as methods designed for one setting make assumptions that break down when the other constraint is simultaneously imposed. In this work, we propose the first comprehensive benchmark for VAD on the edge in the continual learning scenario, evaluating seven VAD models across three lightweight backbone architectures. Furthermore, we propose Tiny-Dinomaly, a lightweight adaptation of the Dinomaly model built on the DINO foundation model that achieves 13x smaller memory footprint and 20x lower computational cost while improving Pixel F1 by 5 percentage points. Finally, we introduce targeted modifications to PatchCore and PaDiM to improve their efficiency in the continual learning setting.