FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3210061328> ?p ?o ?g. }

• W3210061328 endingPage "1197" @default.
• W3210061328 startingPage "1186" @default.
• W3210061328 abstract "Learning with noise is a practically challenging problem in deep face recognition. Despite the success of large margin softmax loss functions, these methods are designed for clean face databases. Considering the inevitable noise in the large scale databases, we first analyze the performance of noise in the training databases. For noise-robust deep face recognition, we propose a dynamic training data dropout (DTDD) method to dynamically filter the noise in the training database and gradually form a stable refined database for model learning. Specifically, we leverage the information provided by the model predictions of accumulated training epochs, which can distinguish regular samples and noise effectively and accurately. The proposed DTDD method is easy and stable for implementation, and can be combined with existing state-of-the-art loss functions and network architectures. Extensive experiments on CASIA-WebFace, VGGFace2, and MS-Celeb-1 M databases empirically demonstrate that our proposed method can robustly train deep face recognition models in the presence of label noise and low quality images." @default.
• W3210061328 created "2021-11-08" @default.
• W3210061328 creator A5013433372 @default.
• W3210061328 creator A5025374137 @default.
• W3210061328 creator A5025452586 @default.
• W3210061328 creator A5028920827 @default.
• W3210061328 creator A5032465074 @default.
• W3210061328 creator A5061256224 @default.
• W3210061328 creator A5063224243 @default.
• W3210061328 date "2022-01-01" @default.
• W3210061328 modified "2023-10-14" @default.
• W3210061328 title "Dynamic Training Data Dropout for Robust Deep Face Recognition" @default.
• W3210061328 cites W1949778830 @default.
• W3210061328 cites W2019464758 @default.
• W3210061328 cites W2024922353 @default.
• W3210061328 cites W2145287260 @default.
• W3210061328 cites W2167460663 @default.
• W3210061328 cites W2194775991 @default.
• W3210061328 cites W2325939864 @default.
• W3210061328 cites W2404498690 @default.
• W3210061328 cites W2515770085 @default.
• W3210061328 cites W2520774990 @default.
• W3210061328 cites W2557864411 @default.
• W3210061328 cites W2618530766 @default.
• W3210061328 cites W2663800299 @default.
• W3210061328 cites W2737608545 @default.
• W3210061328 cites W2752782242 @default.
• W3210061328 cites W2781292787 @default.
• W3210061328 cites W2797558164 @default.
• W3210061328 cites W2799041689 @default.
• W3210061328 cites W2871667416 @default.
• W3210061328 cites W2882991827 @default.
• W3210061328 cites W2887500939 @default.
• W3210061328 cites W2891950293 @default.
• W3210061328 cites W2901505625 @default.
• W3210061328 cites W2949007385 @default.
• W3210061328 cites W2955488837 @default.
• W3210061328 cites W2962895364 @default.
• W3210061328 cites W2962898354 @default.
• W3210061328 cites W2963227127 @default.
• W3210061328 cites W2963460857 @default.
• W3210061328 cites W2963466847 @default.
• W3210061328 cites W2963559058 @default.
• W3210061328 cites W2963671154 @default.
• W3210061328 cites W2963814162 @default.
• W3210061328 cites W2963839617 @default.
• W3210061328 cites W2964155802 @default.
• W3210061328 cites W2964274690 @default.
• W3210061328 cites W2964292098 @default.
• W3210061328 cites W2967637014 @default.
• W3210061328 cites W2969985801 @default.
• W3210061328 cites W2970084480 @default.
• W3210061328 cites W2974328324 @default.
• W3210061328 cites W2984006054 @default.
• W3210061328 cites W2985817549 @default.
• W3210061328 cites W2988966271 @default.
• W3210061328 cites W2996060033 @default.
• W3210061328 cites W2997199946 @default.
• W3210061328 cites W2997312573 @default.
• W3210061328 cites W3000996870 @default.
• W3210061328 cites W3034882062 @default.
• W3210061328 cites W3035376925 @default.
• W3210061328 cites W3094843939 @default.
• W3210061328 cites W3099206234 @default.
• W3210061328 cites W3101227480 @default.
• W3210061328 cites W3141039124 @default.
• W3210061328 doi "https://doi.org/10.1109/tmm.2021.3123478" @default.
• W3210061328 hasPublicationYear "2022" @default.
• W3210061328 type Work @default.
• W3210061328 sameAs 3210061328 @default.
• W3210061328 citedByCount "6" @default.
• W3210061328 countsByYear W32100613282022 @default.
• W3210061328 countsByYear W32100613282023 @default.
• W3210061328 crossrefType "journal-article" @default.
• W3210061328 hasAuthorship W3210061328A5013433372 @default.
• W3210061328 hasAuthorship W3210061328A5025374137 @default.
• W3210061328 hasAuthorship W3210061328A5025452586 @default.
• W3210061328 hasAuthorship W3210061328A5028920827 @default.
• W3210061328 hasAuthorship W3210061328A5032465074 @default.
• W3210061328 hasAuthorship W3210061328A5061256224 @default.
• W3210061328 hasAuthorship W3210061328A5063224243 @default.
• W3210061328 hasConcept C108583219 @default.
• W3210061328 hasConcept C115961682 @default.
• W3210061328 hasConcept C119857082 @default.
• W3210061328 hasConcept C124101348 @default.
• W3210061328 hasConcept C153083717 @default.
• W3210061328 hasConcept C153180895 @default.
• W3210061328 hasConcept C154945302 @default.
• W3210061328 hasConcept C163294075 @default.
• W3210061328 hasConcept C188441871 @default.
• W3210061328 hasConcept C2776145597 @default.
• W3210061328 hasConcept C28490314 @default.
• W3210061328 hasConcept C29265498 @default.
• W3210061328 hasConcept C31510193 @default.
• W3210061328 hasConcept C41008148 @default.
• W3210061328 hasConcept C95623464 @default.
• W3210061328 hasConcept C99498987 @default.
• W3210061328 hasConceptScore W3210061328C108583219 @default.

• next