EdgeZSAD: Practical Zero-Shot Anomaly Detection on Edge Devices

2026-06-15 · Taewan Cho, Andrew Jaeyong Choi arxiv

Industrial inspection needs zero-shot anomaly detection (ZSAD) that remains useful under edge deployment constraints. Recent methods often rely on ViT-L foundation backbones (~300M parameters), which exceed the memory and operator budget of typical embedded hardware. We study this regime through EdgeZSAD, a compact reference system built around a TinyViT-21M-512 backbone, an asymmetric global-local readout (EdgeGLR), and a reproducible source-side training recipe (Real-IAD-DR). We train a single checkpoint in a source-trained, target-unseen protocol and evaluate it across six industrial benchmarks. Across three independent runs, the resulting model reaches an average image AUROC of 91.6 on MVTec-AD and 88.2 on VisA, while remaining directly deployable on Jetson Orin Nano Super (TensorRT FP16) and RB5 Gen2 (QNN GPU FP16). Across the six device-rescored benchmarks, image-AUROC drift stays below 0.2 points, indicating that the exported graph preserves host-side ranking behavior in the evaluated deployment setting.

📄 PDF Abstract BibTeX arXiv:2606.16119

Code (0)

등록된 구현이 없습니다.

Tasks

Anomaly Detection

Results from the Paper

Rank	Task	Dataset	Model	Metrics
#35	Anomaly Detection	VisA	EdgeZSAD	Detection AUROC: 88.2

A Unified Reasoning Framework for Holistic Zero-Shot Video Anomaly Analysis

2025-11-02 · Dongheng Lin, Mengxue Qu, Kunyang Han, Jianbo Jiao 외 arxiv

Most video-anomaly research stops at frame-wise detection, offering little insight into why an event is abnormal, typically outputting only frame-wise anomaly scores without spatial or semantic context. Recent video anom…

Video Anomaly Detection

Towards Zero-shot Point Cloud Anomaly Detection: A Multi-View Projection Framework

2024-09-20 · Yuqi Cheng, Yunkang Cao, Guoyang Xie, Zhichao Lu 외

Detecting anomalies within point clouds is crucial for various industrial applications, but traditional unsupervised methods face challenges due to data acquisition costs, early-stage production constraints, and limited …

Anomaly Detection Specificity zero-shot anomaly detection

FADE: Few-shot/zero-shot Anomaly Detection Engine using Large Vision-Language Model

2024-08-31 · Yuanwei Li, Elizaveta Ivanova, Martins Bruveris

Automatic image anomaly detection is important for quality inspection in the manufacturing industry. The usual unsupervised anomaly detection approach is to train a model for each object class using a dataset of normal s…

Anomaly Detection Anomaly Segmentation Language Modeling Language Modelling+2

LLM-Powered Text-Attributed Graph Anomaly Detection via Retrieval-Augmented Reasoning

2025-11-16 · Haoyan Xu, Ruizhi Qian, Zhengtao Yao, Ziyi Liu 외 arxiv

Anomaly detection on attributed graphs plays an essential role in applications such as fraud detection, intrusion monitoring, and misinformation analysis. However, text-attributed graphs (TAGs), in which node information…

Graph Anomaly Detection Prompt Engineering Fraud Detection

AnomalyVFM -- Transforming Vision Foundation Models into Zero-Shot Anomaly Detectors

2026-01-28 · Matic Fučka, Vitjan Zavrtanik, Danijel Skočaj arxiv

Zero-shot anomaly detection aims to detect and localise abnormal regions in the image without access to any in-domain training images. While recent approaches leverage vision-language models (VLMs), such as CLIP, to tran…