Matches in SemOpenAlex for { <https://semopenalex.org/work/W2949069452> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W2949069452 abstract "Part-based image classification aims at representing categories by small sets of learned discriminative parts, upon which an image representation is built. Considered as a promising avenue a decade ago, this direction has been neglected since the advent of deep neural networks. In this context, this paper brings two contributions: first, it shows that despite the recent success of end-to-end holistic models, explicit part learning can boosts classification performance. Second, this work proceeds one step further than recent part-based models (PBM), focusing on how to learn parts without using any labeled data. Instead of learning a set of parts per class, as generally done in the PBM literature, the proposed approach both constructs a partition of a given set of images into visually similar groups, and subsequently learn a set of discriminative parts per group in a fully unsupervised fashion. This strategy opens the door to the use of PBM in new applications for which the notion of image categories is irrelevant, such as instance-based image retrieval, for example. We experimentally show that our learned parts can help building efficient image representations, for classification as well as for indexing tasks, resulting in performance superior to holistic state-of-the art Deep Convolutional Neural Networks (DCNN) encoding." @default.
- W2949069452 created "2019-06-27" @default.
- W2949069452 creator A5001375778 @default.
- W2949069452 creator A5030078277 @default.
- W2949069452 creator A5047619121 @default.
- W2949069452 creator A5057169414 @default.
- W2949069452 date "2017-04-01" @default.
- W2949069452 modified "2023-09-27" @default.
- W2949069452 title "Unsupervised part learning for visual recognition" @default.
- W2949069452 cites W1169995941 @default.
- W2949069452 cites W1686810756 @default.
- W2949069452 cites W1976591483 @default.
- W2949069452 cites W1979572481 @default.
- W2949069452 cites W1995444699 @default.
- W2949069452 cites W2014788385 @default.
- W2949069452 cites W2044180021 @default.
- W2949069452 cites W2074800359 @default.
- W2949069452 cites W2115628259 @default.
- W2949069452 cites W2116045373 @default.
- W2949069452 cites W2121348293 @default.
- W2949069452 cites W2129305389 @default.
- W2949069452 cites W2141362318 @default.
- W2949069452 cites W2148809531 @default.
- W2949069452 cites W2150307973 @default.
- W2949069452 cites W2151625907 @default.
- W2949069452 cites W2152161678 @default.
- W2949069452 cites W2164488048 @default.
- W2949069452 cites W2167828171 @default.
- W2949069452 cites W2168356304 @default.
- W2949069452 cites W2194775991 @default.
- W2949069452 cites W2949368876 @default.
- W2949069452 cites W2951702175 @default.
- W2949069452 cites W2951776594 @default.
- W2949069452 cites W2952271367 @default.
- W2949069452 cites W2953270581 @default.
- W2949069452 cites W2963166708 @default.
- W2949069452 hasPublicationYear "2017" @default.
- W2949069452 type Work @default.
- W2949069452 sameAs 2949069452 @default.
- W2949069452 citedByCount "0" @default.
- W2949069452 crossrefType "posted-content" @default.
- W2949069452 hasAuthorship W2949069452A5001375778 @default.
- W2949069452 hasAuthorship W2949069452A5030078277 @default.
- W2949069452 hasAuthorship W2949069452A5047619121 @default.
- W2949069452 hasAuthorship W2949069452A5057169414 @default.
- W2949069452 hasBestOaLocation W29490694521 @default.
- W2949069452 hasConcept C153180895 @default.
- W2949069452 hasConcept C154945302 @default.
- W2949069452 hasConcept C15744967 @default.
- W2949069452 hasConcept C41008148 @default.
- W2949069452 hasConcept C8038995 @default.
- W2949069452 hasConceptScore W2949069452C153180895 @default.
- W2949069452 hasConceptScore W2949069452C154945302 @default.
- W2949069452 hasConceptScore W2949069452C15744967 @default.
- W2949069452 hasConceptScore W2949069452C41008148 @default.
- W2949069452 hasConceptScore W2949069452C8038995 @default.
- W2949069452 hasLocation W29490694521 @default.
- W2949069452 hasLocation W29490694522 @default.
- W2949069452 hasLocation W29490694523 @default.
- W2949069452 hasLocation W29490694524 @default.
- W2949069452 hasLocation W29490694525 @default.
- W2949069452 hasOpenAccess W2949069452 @default.
- W2949069452 hasPrimaryLocation W29490694521 @default.
- W2949069452 hasRelatedWork W1548481688 @default.
- W2949069452 hasRelatedWork W2012485643 @default.
- W2949069452 hasRelatedWork W2026253357 @default.
- W2949069452 hasRelatedWork W2292254049 @default.
- W2949069452 hasRelatedWork W2592385986 @default.
- W2949069452 hasRelatedWork W2608244370 @default.
- W2949069452 hasRelatedWork W2695951553 @default.
- W2949069452 hasRelatedWork W2775464024 @default.
- W2949069452 hasRelatedWork W2899683012 @default.
- W2949069452 hasRelatedWork W3108696707 @default.
- W2949069452 isParatext "false" @default.
- W2949069452 isRetracted "false" @default.
- W2949069452 magId "2949069452" @default.
- W2949069452 workType "article" @default.