Matches in SemOpenAlex for { <https://semopenalex.org/work/W2891951760> ?p ?o ?g. }
- W2891951760 endingPage "454" @default.
- W2891951760 startingPage "438" @default.
- W2891951760 abstract "Fine-grained classification is challenging due to the difficulty of finding discriminative features. Finding those subtle traits that fully characterize the object is not straightforward. To handle this circumstance, we propose a novel self-supervision mechanism to effectively localize informative regions without the need of bounding-box/part annotations. Our model, termed NTS-Net for Navigator-Teacher-Scrutinizer Network, consists of a Navigator agent, a Teacher agent and a Scrutinizer agent. In consideration of intrinsic consistency between informativeness of the regions and their probability being ground-truth class, we design a novel training paradigm, which enables Navigator to detect most informative regions under the guidance from Teacher. After that, the Scrutinizer scrutinizes the proposed regions from Navigator and makes predictions. Our model can be viewed as a multi-agent cooperation, wherein agents benefit from each other, and make progress together. NTS-Net can be trained end-to-end, while provides accurate fine-grained classification predictions as well as highly informative regions during inference. We achieve state-of-the-art performance in extensive benchmark datasets." @default.
- W2891951760 created "2018-09-27" @default.
- W2891951760 creator A5003074251 @default.
- W2891951760 creator A5046454314 @default.
- W2891951760 creator A5055067260 @default.
- W2891951760 creator A5055723755 @default.
- W2891951760 creator A5078250869 @default.
- W2891951760 creator A5086976460 @default.
- W2891951760 date "2018-01-01" @default.
- W2891951760 modified "2023-10-14" @default.
- W2891951760 title "Learning to Navigate for Fine-Grained Classification" @default.
- W2891951760 cites W1531492214 @default.
- W2891951760 cites W1555385401 @default.
- W2891951760 cites W1898560071 @default.
- W2891951760 cites W1903029394 @default.
- W2891951760 cites W1967632700 @default.
- W2891951760 cites W1980526845 @default.
- W2891951760 cites W1984514441 @default.
- W2891951760 cites W1991367009 @default.
- W2891951760 cites W1995543189 @default.
- W2891951760 cites W2009303086 @default.
- W2891951760 cites W2046382188 @default.
- W2891951760 cites W2088049833 @default.
- W2891951760 cites W2091158010 @default.
- W2891951760 cites W2102605133 @default.
- W2891951760 cites W2104657103 @default.
- W2891951760 cites W2108862644 @default.
- W2891951760 cites W2109255472 @default.
- W2891951760 cites W2117539524 @default.
- W2891951760 cites W2118696714 @default.
- W2891951760 cites W2135706578 @default.
- W2891951760 cites W2138011018 @default.
- W2891951760 cites W2143166796 @default.
- W2891951760 cites W2143331230 @default.
- W2891951760 cites W2149427297 @default.
- W2891951760 cites W2151103935 @default.
- W2891951760 cites W2194775991 @default.
- W2891951760 cites W2207849498 @default.
- W2891951760 cites W2479109623 @default.
- W2891951760 cites W2565639579 @default.
- W2891951760 cites W2604195031 @default.
- W2891951760 cites W2613763509 @default.
- W2891951760 cites W2741910023 @default.
- W2891951760 cites W2773003563 @default.
- W2891951760 cites W2780838211 @default.
- W2891951760 cites W2963037989 @default.
- W2891951760 cites W2963725249 @default.
- W2891951760 cites W3106250896 @default.
- W2891951760 cites W3124951096 @default.
- W2891951760 cites W56385144 @default.
- W2891951760 doi "https://doi.org/10.1007/978-3-030-01264-9_26" @default.
- W2891951760 hasPublicationYear "2018" @default.
- W2891951760 type Work @default.
- W2891951760 sameAs 2891951760 @default.
- W2891951760 citedByCount "297" @default.
- W2891951760 countsByYear W28919517602019 @default.
- W2891951760 countsByYear W28919517602020 @default.
- W2891951760 countsByYear W28919517602021 @default.
- W2891951760 countsByYear W28919517602022 @default.
- W2891951760 countsByYear W28919517602023 @default.
- W2891951760 crossrefType "book-chapter" @default.
- W2891951760 hasAuthorship W2891951760A5003074251 @default.
- W2891951760 hasAuthorship W2891951760A5046454314 @default.
- W2891951760 hasAuthorship W2891951760A5055067260 @default.
- W2891951760 hasAuthorship W2891951760A5055723755 @default.
- W2891951760 hasAuthorship W2891951760A5078250869 @default.
- W2891951760 hasAuthorship W2891951760A5086976460 @default.
- W2891951760 hasBestOaLocation W28919517602 @default.
- W2891951760 hasConcept C115961682 @default.
- W2891951760 hasConcept C119857082 @default.
- W2891951760 hasConcept C13280743 @default.
- W2891951760 hasConcept C146849305 @default.
- W2891951760 hasConcept C147037132 @default.
- W2891951760 hasConcept C154945302 @default.
- W2891951760 hasConcept C185798385 @default.
- W2891951760 hasConcept C205649164 @default.
- W2891951760 hasConcept C2776214188 @default.
- W2891951760 hasConcept C2776436953 @default.
- W2891951760 hasConcept C2777212361 @default.
- W2891951760 hasConcept C2781238097 @default.
- W2891951760 hasConcept C41008148 @default.
- W2891951760 hasConcept C63584917 @default.
- W2891951760 hasConcept C774472 @default.
- W2891951760 hasConcept C97931131 @default.
- W2891951760 hasConceptScore W2891951760C115961682 @default.
- W2891951760 hasConceptScore W2891951760C119857082 @default.
- W2891951760 hasConceptScore W2891951760C13280743 @default.
- W2891951760 hasConceptScore W2891951760C146849305 @default.
- W2891951760 hasConceptScore W2891951760C147037132 @default.
- W2891951760 hasConceptScore W2891951760C154945302 @default.
- W2891951760 hasConceptScore W2891951760C185798385 @default.
- W2891951760 hasConceptScore W2891951760C205649164 @default.
- W2891951760 hasConceptScore W2891951760C2776214188 @default.
- W2891951760 hasConceptScore W2891951760C2776436953 @default.
- W2891951760 hasConceptScore W2891951760C2777212361 @default.
- W2891951760 hasConceptScore W2891951760C2781238097 @default.
- W2891951760 hasConceptScore W2891951760C41008148 @default.
- W2891951760 hasConceptScore W2891951760C63584917 @default.