Leveraging Unsupervised Learning for Cost-Effective Visual Anomaly Detection

In three linesVisual anomaly detection system using unsupervised learning on Raspberry Pi. Training and inference in 90 seconds with 10 normal images, F1 score >0.95. Deployment via Anomalib and openVINO for SMEs.

Summary generated by Claude — human-verified