KairosAD: A SAM-Based Model for Industrial Anomaly Detection on Embedded Devices
This addresses the need for practical, efficient anomaly detection on embedded devices in small and medium enterprises, though it is incremental as it adapts existing models for specific deployment constraints.
The paper tackles the problem of deploying anomaly detection models on resource-constrained embedded devices for industrial manufacturing by proposing KairosAD, which uses MobileSAM to achieve 78% fewer parameters and 4x faster inference time while maintaining comparable AUROC performance on datasets like MVTec-AD and ViSA.
In the era of intelligent manufacturing, anomaly detection has become essential for maintaining quality control on modern production lines. However, while many existing models show promising performance, they are often too large, computationally demanding, and impractical to deploy on resource-constrained embedded devices that can be easily installed on the production lines of Small and Medium Enterprises (SMEs). To bridge this gap, we present KairosAD, a novel supervised approach that uses the power of the Mobile Segment Anything Model (MobileSAM) for image-based anomaly detection. KairosAD has been evaluated on the two well-known industrial anomaly detection datasets, i.e., MVTec-AD and ViSA. The results show that KairosAD requires 78% fewer parameters and boasts a 4x faster inference time compared to the leading state-of-the-art model, while maintaining comparable AUROC performance. We deployed KairosAD on two embedded devices, the NVIDIA Jetson NX, and the NVIDIA Jetson AGX. Finally, KairosAD was successfully installed and tested on the real production line of the Industrial Computer Engineering Laboratory (ICE Lab) at the University of Verona. The code is available at https://github.com/intelligolabs/KairosAD.