FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312274522> ?p ?o ?g. }

• W4312274522 endingPage "2189" @default.
• W4312274522 startingPage "2176" @default.
• W4312274522 abstract "Recently, much attention has been paid to segmenting subtle unknown defect regions by knowledge distillation in an unsupervised setting. Most previous studies concentrated on guiding the student network to learn the same representations on the normality, neglecting the different behaviors of the abnormality. This leads to a high probability of false detection of subtle defects. To address such an issue, we propose to push representations on abnormal areas of the teacher and student network as far as possible while pulling representations on normal areas as close as possible. Based on this idea, we design an efficient teacher-student model for anomaly detection and localization, which maximizes pixel-wise discrepancies for anomalous regions approximated by data augmentation and simultaneously minimizes discrepancies for pixel-wise normal regions between these two networks. The explicit differential knowledge distillation enlarges the margin between normal representations and abnormal ones in favour of discriminating them. Then, the appropriate small student network is not only efficient, but more importantly, helps inhibit the generalization ability of anomalous patterns when learning normal patterns, facilitating the precise decision boundary. The experimental results on the MVTec AD, Fashion-MNIST, and CIFAR-10 datasets demonstrate that our proposed method achieves better performance than current state-of-the-art (SOTA) approaches. Especially, For the MVTec AD dataset with high resolution images, we achieve 98.1 AUROC% and 93.6 AUPRO% in anomaly localization, outperforming knowledge distillation based SOTA methods by 1.1 AUROC% and 1.5 AUPRO% with a lightweight model." @default.
• W4312274522 created "2023-01-04" @default.
• W4312274522 creator A5021220108 @default.
• W4312274522 creator A5021637761 @default.
• W4312274522 creator A5067320802 @default.
• W4312274522 creator A5068195118 @default.
• W4312274522 creator A5082733824 @default.
• W4312274522 date "2023-05-01" @default.
• W4312274522 modified "2023-10-17" @default.
• W4312274522 title "Pull & Push: Leveraging Differential Knowledge Distillation for Efficient Unsupervised Anomaly Detection and Localization" @default.
• W4312274522 cites W1970088130 @default.
• W4312274522 cites W2110934250 @default.
• W4312274522 cites W2125276050 @default.
• W4312274522 cites W2133665775 @default.
• W4312274522 cites W2138621090 @default.
• W4312274522 cites W2194775991 @default.
• W4312274522 cites W2524712492 @default.
• W4312274522 cites W2560647685 @default.
• W4312274522 cites W2599354622 @default.
• W4312274522 cites W2809705434 @default.
• W4312274522 cites W2925312408 @default.
• W4312274522 cites W2948982773 @default.
• W4312274522 cites W2963070890 @default.
• W4312274522 cites W2963795951 @default.
• W4312274522 cites W2987228832 @default.
• W4312274522 cites W2998277278 @default.
• W4312274522 cites W3034314048 @default.
• W4312274522 cites W3034406766 @default.
• W4312274522 cites W3034781633 @default.
• W4312274522 cites W3035524453 @default.
• W4312274522 cites W3092704883 @default.
• W4312274522 cites W3108027406 @default.
• W4312274522 cites W3109715690 @default.
• W4312274522 cites W3118895125 @default.
• W4312274522 cites W3135550350 @default.
• W4312274522 cites W3147184966 @default.
• W4312274522 cites W3148545080 @default.
• W4312274522 cites W3160074056 @default.
• W4312274522 cites W3160366495 @default.
• W4312274522 cites W3166166117 @default.
• W4312274522 cites W3169077988 @default.
• W4312274522 cites W3169651898 @default.
• W4312274522 cites W3175716777 @default.
• W4312274522 cites W3183588514 @default.
• W4312274522 cites W3195417828 @default.
• W4312274522 cites W4210446097 @default.
• W4312274522 cites W4281560842 @default.
• W4312274522 cites W4290964450 @default.
• W4312274522 cites W4312772600 @default.
• W4312274522 doi "https://doi.org/10.1109/tcsvt.2022.3218587" @default.
• W4312274522 hasPublicationYear "2023" @default.
• W4312274522 type Work @default.
• W4312274522 citedByCount "1" @default.
• W4312274522 countsByYear W43122745222023 @default.
• W4312274522 crossrefType "journal-article" @default.
• W4312274522 hasAuthorship W4312274522A5021220108 @default.
• W4312274522 hasAuthorship W4312274522A5021637761 @default.
• W4312274522 hasAuthorship W4312274522A5067320802 @default.
• W4312274522 hasAuthorship W4312274522A5068195118 @default.
• W4312274522 hasAuthorship W4312274522A5082733824 @default.
• W4312274522 hasConcept C119857082 @default.
• W4312274522 hasConcept C121332964 @default.
• W4312274522 hasConcept C12267149 @default.
• W4312274522 hasConcept C12997251 @default.
• W4312274522 hasConcept C134306372 @default.
• W4312274522 hasConcept C153180895 @default.
• W4312274522 hasConcept C154945302 @default.
• W4312274522 hasConcept C15744967 @default.
• W4312274522 hasConcept C160633673 @default.
• W4312274522 hasConcept C177148314 @default.
• W4312274522 hasConcept C26873012 @default.
• W4312274522 hasConcept C33923547 @default.
• W4312274522 hasConcept C41008148 @default.
• W4312274522 hasConcept C42023084 @default.
• W4312274522 hasConcept C50965678 @default.
• W4312274522 hasConcept C739882 @default.
• W4312274522 hasConcept C774472 @default.
• W4312274522 hasConcept C77805123 @default.
• W4312274522 hasConcept C8038995 @default.
• W4312274522 hasConceptScore W4312274522C119857082 @default.
• W4312274522 hasConceptScore W4312274522C121332964 @default.
• W4312274522 hasConceptScore W4312274522C12267149 @default.
• W4312274522 hasConceptScore W4312274522C12997251 @default.
• W4312274522 hasConceptScore W4312274522C134306372 @default.
• W4312274522 hasConceptScore W4312274522C153180895 @default.
• W4312274522 hasConceptScore W4312274522C154945302 @default.
• W4312274522 hasConceptScore W4312274522C15744967 @default.
• W4312274522 hasConceptScore W4312274522C160633673 @default.
• W4312274522 hasConceptScore W4312274522C177148314 @default.
• W4312274522 hasConceptScore W4312274522C26873012 @default.
• W4312274522 hasConceptScore W4312274522C33923547 @default.
• W4312274522 hasConceptScore W4312274522C41008148 @default.
• W4312274522 hasConceptScore W4312274522C42023084 @default.
• W4312274522 hasConceptScore W4312274522C50965678 @default.
• W4312274522 hasConceptScore W4312274522C739882 @default.
• W4312274522 hasConceptScore W4312274522C774472 @default.
• W4312274522 hasConceptScore W4312274522C77805123 @default.
• W4312274522 hasConceptScore W4312274522C8038995 @default.

• next