PaperWACV 2024 (arXiv 2023) · 2023
EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latencies
1page built on it
3more pages cite it
2narratives
0registry papers cite it
In the Atlas
ModelPrimary source of
EfficientAD
Millisecond-latency one-class anomaly detection combining a loss-induced-asymmetric student-teacher branch for structural defects with a feature-space autoencoder branch for logical defects, running 2.2 ms (S) / 4.5 ms (M) per image on an RTX A6000.
Cited as a reference by 3 pages, grouped by domain
Anomaly detection 3
Appears in 2 narratives
Lineage
Papers in the Atlas registry only · one dot per paperBuilt upon by · 0
Builds on · 6
2019MVTec AD — A Comprehensive Real-World Dataset for Unsupervised Anomaly DetectionBergmann, Fauser et al.2020Uninformed Students: Student-Teacher Anomaly Detection With Discriminative Latent EmbeddingsBergmann, Fauser et al.2022Beyond Dents and Scratches: Logical Constraints in Unsupervised Anomaly Detection and LocalizationBergmann, Batzner et al.2022Towards Total Recall in Industrial Anomaly DetectionRoth, Pemula et al.2022SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and SegmentationZou, Jeong et al.2023Asymmetric Student-Teacher Networks for Industrial Anomaly DetectionRudolph, Wehrbein et al.Built upon by · 0