Matches in SemOpenAlex for { <https://semopenalex.org/work/W2912106445> ?p ?o ?g. }
- W2912106445 abstract "In this paper, we study the challenging unconstrained set-based face recognition problem where each subject face is instantiated by a set of media (images and videos) instead of a single image. Naively aggregating information from all the media within a set would suffer from the large intra-set variance caused by heterogeneous factors (e.g., varying media modalities, poses and illuminations) and fail to learn discriminative face representations. A novel Multi-Prototype Network (MPNet) model is thus proposed to learn multiple prototype face representations adaptively from the media sets. Each learned prototype is representative for the subject face under certain condition in terms of pose, illumination and media modality. Instead of handcrafting the set partition for prototype learning, MPNet introduces a Dense SubGraph (DSG) learning sub-net that implicitly untangles inconsistent media and learns a number of representative prototypes. Qualitative and quantitative experiments clearly demonstrate superiority of the proposed model over state-of-the-arts." @default.
- W2912106445 created "2019-02-21" @default.
- W2912106445 creator A5008135430 @default.
- W2912106445 creator A5011866398 @default.
- W2912106445 creator A5013979705 @default.
- W2912106445 creator A5033884558 @default.
- W2912106445 creator A5039765869 @default.
- W2912106445 creator A5057739949 @default.
- W2912106445 creator A5076090670 @default.
- W2912106445 creator A5078587654 @default.
- W2912106445 creator A5087073100 @default.
- W2912106445 date "2019-02-13" @default.
- W2912106445 modified "2023-09-23" @default.
- W2912106445 title "Multi-Prototype Networks for Unconstrained Set-based Face Recognition" @default.
- W2912106445 cites W1665214252 @default.
- W2912106445 cites W1677182931 @default.
- W2912106445 cites W1686810756 @default.
- W2912106445 cites W1782590233 @default.
- W2912106445 cites W1895390915 @default.
- W2912106445 cites W1921147789 @default.
- W2912106445 cites W1949778830 @default.
- W2912106445 cites W1950843348 @default.
- W2912106445 cites W2019464758 @default.
- W2912106445 cites W2066650804 @default.
- W2912106445 cites W2140609507 @default.
- W2912106445 cites W2144172034 @default.
- W2912106445 cites W2145287260 @default.
- W2912106445 cites W2155893237 @default.
- W2912106445 cites W2187089797 @default.
- W2912106445 cites W2194775991 @default.
- W2912106445 cites W2271887239 @default.
- W2912106445 cites W2275811363 @default.
- W2912106445 cites W2307523691 @default.
- W2912106445 cites W2325939864 @default.
- W2912106445 cites W2335787378 @default.
- W2912106445 cites W2441198140 @default.
- W2912106445 cites W2520774990 @default.
- W2912106445 cites W2522992079 @default.
- W2912106445 cites W2524108254 @default.
- W2912106445 cites W2526639894 @default.
- W2912106445 cites W2619432459 @default.
- W2912106445 cites W2753407461 @default.
- W2912106445 cites W2784874046 @default.
- W2912106445 cites W2798807671 @default.
- W2912106445 cites W2798951992 @default.
- W2912106445 cites W2808174311 @default.
- W2912106445 cites W2871667416 @default.
- W2912106445 cites W2884928667 @default.
- W2912106445 cites W2951552869 @default.
- W2912106445 cites W2952198537 @default.
- W2912106445 cites W2962725301 @default.
- W2912106445 cites W2962898354 @default.
- W2912106445 cites W2963466847 @default.
- W2912106445 cites W2963559058 @default.
- W2912106445 cites W2963570419 @default.
- W2912106445 cites W2963839617 @default.
- W2912106445 cites W3099206234 @default.
- W2912106445 cites W3099887033 @default.
- W2912106445 cites W3101036461 @default.
- W2912106445 cites W3103152812 @default.
- W2912106445 cites W975491072 @default.
- W2912106445 doi "https://doi.org/10.48550/arxiv.1902.04755" @default.
- W2912106445 hasPublicationYear "2019" @default.
- W2912106445 type Work @default.
- W2912106445 sameAs 2912106445 @default.
- W2912106445 citedByCount "2" @default.
- W2912106445 countsByYear W29121064452020 @default.
- W2912106445 crossrefType "posted-content" @default.
- W2912106445 hasAuthorship W2912106445A5008135430 @default.
- W2912106445 hasAuthorship W2912106445A5011866398 @default.
- W2912106445 hasAuthorship W2912106445A5013979705 @default.
- W2912106445 hasAuthorship W2912106445A5033884558 @default.
- W2912106445 hasAuthorship W2912106445A5039765869 @default.
- W2912106445 hasAuthorship W2912106445A5057739949 @default.
- W2912106445 hasAuthorship W2912106445A5076090670 @default.
- W2912106445 hasAuthorship W2912106445A5078587654 @default.
- W2912106445 hasAuthorship W2912106445A5087073100 @default.
- W2912106445 hasBestOaLocation W29121064451 @default.
- W2912106445 hasConcept C114614502 @default.
- W2912106445 hasConcept C115961682 @default.
- W2912106445 hasConcept C119857082 @default.
- W2912106445 hasConcept C144024400 @default.
- W2912106445 hasConcept C153180895 @default.
- W2912106445 hasConcept C154945302 @default.
- W2912106445 hasConcept C161191863 @default.
- W2912106445 hasConcept C177264268 @default.
- W2912106445 hasConcept C199360897 @default.
- W2912106445 hasConcept C2777855551 @default.
- W2912106445 hasConcept C2779304628 @default.
- W2912106445 hasConcept C2780226545 @default.
- W2912106445 hasConcept C31510193 @default.
- W2912106445 hasConcept C31972630 @default.
- W2912106445 hasConcept C33923547 @default.
- W2912106445 hasConcept C36289849 @default.
- W2912106445 hasConcept C41008148 @default.
- W2912106445 hasConcept C42812 @default.
- W2912106445 hasConcept C97931131 @default.
- W2912106445 hasConceptScore W2912106445C114614502 @default.
- W2912106445 hasConceptScore W2912106445C115961682 @default.
- W2912106445 hasConceptScore W2912106445C119857082 @default.