An experimental evaluation review of recent methods using common dataset splits, preprocessing procedures, and evaluation metrics, while retaining method-specific backbones, input resolutions, and optimization schedules shows that visual-only methods achieve the strongest peak performance on standard benchmarks, but becomes less consistent on MPDD, under limited-data benchmark settings, and on the real EV dataset.
Abstract
Industrial anomaly detection remains challenging in manufacturing quality inspection because defects are often subtle, spatially sensitive, and costly to annotate. These constraints have motivated a recent shift from visual-only models toward vision–language approaches that can represent anomalies more flexibly. However, it remains unclear whether this shift consistently improves detection accuracy and deployment practicality under industrial conditions. To address this gap, this paper presents an experimental evaluation review of recent methods using common dataset splits, preprocessing procedures, and evaluation metrics, while retaining method-specific backbones, input resolutions, and optimization schedules. The comparison covers standard benchmarks, limited-data benchmark settings, and the Rotated EV Metal Triangle dataset derived from real electric-vehicle manufacturing data, with further analysis across texture, spatially regular object, and pose-varying object categories. Benchmark comparisons are based on single runs and interpreted descriptively, while EV dataset results are evaluated across repeated runs. Beyond image-level and pixel-level detection performance, the evaluation also considers computational efficiency to assess practical deployment trade-offs. The results show that visual-only methods achieve the strongest peak performance on standard benchmarks, with RD++ providing a favorable accuracy–speed trade-off under its evaluated configuration. However, this advantage becomes less consistent on MPDD, under limited-data benchmark settings, and on the real EV dataset, where pose variation, reduced training availability, and more challenging test conditions affect the reliability of several methods. In these settings, vision–language methods with learnable text alignment become competitive, with RareCLIP ranking among the strongest performers.
Industrial anomaly detection benchmarks have advanced rapidly, yet they remain largely static and therefore underrepresent deployment settings in which inspection systems must be extended from one industrial component to another over time. In our setting, continual industrial anomaly detection is framed as a task-incre...
L. Marcjan, D. Wilk-Kołodziejczyk, Roberto Corizzo· 0 citations
Some industrial defects are geometric and invisible to a color camera, and anomaly detection on point clouds, depth maps and RGB-D images grew up around them. This survey covers 2020 to 21 August 2026, 94 detectors and 17 benchmark datasets, with an executable search protocol tested against the corpus and every collate...
Lars Fichtel, R. Timofte, Dominik Seuss· Italian National Conference...· 0 citations
Industrial anomaly detection is a critical task in automated visual inspection systems, aiming to identify subtle defects without relying on large-scale anomaly annotations. Recently, pretrained vision transformers such as DINO have demonstrated strong feature representation capabilities and have been widely adopted in...
Chao-Qun Wang, Wen-Jing Zhang, Bo Qi et al.· Multimedia Systems· 0 citations
Training-free industrial anomaly localization avoids category-specific model training, but benchmark scores alone provide limited evidence about operational reliability. We present a fixed-protocol study on all eight MVTec AD 2 categories, comparing five variants across three families: three DINO configurations that jo...
Vi-Long Vo, L. Pham, Thai Duy Pham et al.· IEEE Access· 0 citations
Risk-Calibrated Few-Shot Industrial Anomaly Detection Plus (RC-FS-IAD+), afew-normal-support, labeled-calibration-assisted framework that maps anomaly scores to automatic pass, manual review, and automatic rejection is presented.
Ismail Hakki Kinalioglu· Journal of Advanced Research...· 0 citations
A framework in which a classifier, trained on normal images and on synthetic anomalies, scores a test image without comparing it with normal data is proposed, suggesting that clean benchmark scores can hide a strong sensitivity to capture conditions, and that learning the decision from both normal images and rule viola...
Ufuk Efe, A. Temi̇zel· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.