Search papers, labs, and topics across Lattice.
This paper introduces LiZAD, a lightweight zero-shot anomaly detection framework optimized for real-time deployment on edge devices in industrial manufacturing. By integrating the dense visual features of DINOv3 with the efficient text embeddings of MobileCLIP2, LiZAD achieves significant reductions in memory and parameters while maintaining competitive anomaly detection performance. Specifically, it demonstrates a 61.5% reduction in memory usage and a 3.02x speedup in latency compared to six state-of-the-art models, making it suitable for resource-constrained environments.
LiZAD slashes memory and latency requirements for zero-shot anomaly detection by over 60%, making real-time defect detection feasible on edge devices.
In modern high-throughput industrial production lines, product configurations and visual characteristics frequently change, making it impractical to collect and annotate data for every new scenario. This dynamic setting makes Zero-Shot Anomaly Detection (ZSAD) particularly suitable, as it enables defect detection without requiring training on target-specific samples. Although recent ZSAD approaches show promising results, they are computationally intensive and thus unsuitable for deployment on resource-constrained devices. We propose LiZAD: a lightweight framework designed for real-time ZSAD specifically tailored for use on edge devices. The proposed approach pairs the dense and spatially aware visual features of DINOv3, crucial for precise pixel-level localization, with the highly computationally efficient text embeddings of MobileCLIP2. These features are then mapped into a shared latent space via low-memory trainable projection heads. Compared to six state-of-the-art ZSAD models, LiZAD achieves an average memory reduction of 61.5%, a parameter reduction of 74.6%, and a speedup of 3.02x in terms of latency. Despite substantial reductions in computational and memory costs, our approach maintains competitive anomaly detection performance, dropping the average P-AUROC by just 6.4% relative to the best state-of-the-art model across the VisA, BTAD, MPDD, and MVTec-AD datasets. Finally, it is successfully deployed on the NVIDIA Jetson NX and Jetson AGX edge devices 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/LiZAD.
