Efficient Visual Anomaly Detection at the Edge: Enabling Real-Time Industrial Inspection on Resource-Constrained Devices

2026-03-18

AI Summary

This work addresses the challenge of deploying visual anomaly detection in industrial quality inspection on edge devices, where real-time performance and data privacy are critical yet existing methods suffer from excessive memory and computational demands. To this end, we propose PatchCore-Lite and PaDiM-Lite—lightweight variants of PatchCore and PaDiM—specifically optimized for edge deployment. Our approach introduces a compact memory bank via product quantization, a coarse-to-fine two-stage search mechanism, and an approximation of the Mahalanobis distance using diagonal covariance, which reduces it to element-wise operations. Evaluated on the MVTec AD and VisA benchmarks, PatchCore-Lite achieves a 79% reduction in memory usage, while PaDiM-Lite reduces memory by 77% and accelerates inference by 31%, enabling efficient, cloud-independent anomaly detection at the edge.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Adversarial Attacks & Robustness
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Security and Privacy: Large-scale security measurements
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Visual Anomaly Detection (VAD) is essential for industrial quality control, enabling automatic defect detection in manufacturing. In real production lines, VAD systems must satisfy strict real-time and privacy requirements, necessitating a shift from cloud-based processing to local edge deployment. However, processing data locally on edge devices introduces new challenges because edge hardware has limited memory and computational resources. To overcome these limitations, we propose two efficient VAD methods designed for edge deployment: PatchCore-Lite and Padim-Lite, based on the popular PatchCore and PaDiM models. PatchCore-Lite runs first a coarse search on a product-quantized memory bank, then an exact search on a decoded subset. Padim-Lite is sped up using diagonal covariance, turning Mahalanobis distance into efficient element-wise computation. We evaluate our methods on the MVTec AD and VisA benchmarks and show their suitability for edge environments. PatchCore-Lite achieves a remarkable 79% reduction in total memory footprint, while PaDiM-Lite achieves substantial efficiency gains with a 77% reduction in total memory and a 31% decrease in inference time. These results show that VAD can be effectively deployed on edge devices, enabling real-time, private, and cost-efficient industrial inspection.

Problem

• Visual Anomaly Detection
• Edge Computing
• Industrial Inspection
• Resource-Constrained Devices
• Real-Time Processing

Innovation

• Edge Computing
• Visual Anomaly Detection
• Model Compression
• Efficient Inference
• Industrial Inspection