Matches in SemOpenAlex for { <https://semopenalex.org/work/W3167827695> ?p ?o ?g. }
- W3167827695 abstract "Metric-based few-shot fine-grained image classification (FSFGIC) aims to learn a transferable feature embedding network by estimating the similarities between query images and support classes from very few examples. In this work, we propose, for the first time, to introduce the non-linear data projection concept into the design of FSFGIC architecture in order to address the limited sample problem in few-shot learning and at the same time to increase the discriminability of the model for fine-grained image classification. Specifically, we first design a feature re-abstraction embedding network that has the ability to not only obtain the required semantic features for effective metric learning but also re-enhance such features with finer details from input images. Then the descriptors of the query images and the support classes are projected into different non-linear spaces in our proposed similarity metric learning network to learn discriminative projection factors. This design can effectively operate in the challenging and restricted condition of a FSFGIC task for making the distance between the samples within the same class smaller and the distance between samples from different classes larger and for reducing the coupling relationship between samples from different categories. Furthermore, a novel similarity measure based on the proposed non-linear data project is presented for evaluating the relationships of feature information between a query image and a support set. It is worth to note that our proposed architecture can be easily embedded into any episodic training mechanisms for end-to-end training from scratch. Extensive experiments on FSFGIC tasks demonstrate the superiority of the proposed methods over the state-of-the-art benchmarks." @default.
- W3167827695 created "2021-06-22" @default.
- W3167827695 creator A5013741704 @default.
- W3167827695 creator A5037060407 @default.
- W3167827695 creator A5041589060 @default.
- W3167827695 creator A5071549690 @default.
- W3167827695 creator A5088046237 @default.
- W3167827695 date "2021-06-13" @default.
- W3167827695 modified "2023-09-27" @default.
- W3167827695 title "NDPNet: A novel non-linear data projection network for few-shot fine-grained image classification." @default.
- W3167827695 cites W1488195538 @default.
- W3167827695 cites W1536680647 @default.
- W3167827695 cites W1542791059 @default.
- W3167827695 cites W2014158063 @default.
- W3167827695 cites W2126204609 @default.
- W3167827695 cites W2138011018 @default.
- W3167827695 cites W2601450892 @default.
- W3167827695 cites W2604763608 @default.
- W3167827695 cites W2618530766 @default.
- W3167827695 cites W2787501667 @default.
- W3167827695 cites W2787705367 @default.
- W3167827695 cites W2900082762 @default.
- W3167827695 cites W2904218366 @default.
- W3167827695 cites W2957105406 @default.
- W3167827695 cites W2963341924 @default.
- W3167827695 cites W2963551321 @default.
- W3167827695 cites W2964105864 @default.
- W3167827695 cites W2964121744 @default.
- W3167827695 cites W2979689312 @default.
- W3167827695 cites W2997737174 @default.
- W3167827695 cites W3034978645 @default.
- W3167827695 cites W3035163205 @default.
- W3167827695 cites W3091905774 @default.
- W3167827695 cites W639708223 @default.
- W3167827695 cites W14333344 @default.
- W3167827695 hasPublicationYear "2021" @default.
- W3167827695 type Work @default.
- W3167827695 sameAs 3167827695 @default.
- W3167827695 citedByCount "0" @default.
- W3167827695 crossrefType "posted-content" @default.
- W3167827695 hasAuthorship W3167827695A5013741704 @default.
- W3167827695 hasAuthorship W3167827695A5037060407 @default.
- W3167827695 hasAuthorship W3167827695A5041589060 @default.
- W3167827695 hasAuthorship W3167827695A5071549690 @default.
- W3167827695 hasAuthorship W3167827695A5088046237 @default.
- W3167827695 hasConcept C103278499 @default.
- W3167827695 hasConcept C111472728 @default.
- W3167827695 hasConcept C11413529 @default.
- W3167827695 hasConcept C115961682 @default.
- W3167827695 hasConcept C119857082 @default.
- W3167827695 hasConcept C124101348 @default.
- W3167827695 hasConcept C124304363 @default.
- W3167827695 hasConcept C138885662 @default.
- W3167827695 hasConcept C153180895 @default.
- W3167827695 hasConcept C154945302 @default.
- W3167827695 hasConcept C162324750 @default.
- W3167827695 hasConcept C176217482 @default.
- W3167827695 hasConcept C177264268 @default.
- W3167827695 hasConcept C199360897 @default.
- W3167827695 hasConcept C21547014 @default.
- W3167827695 hasConcept C2776401178 @default.
- W3167827695 hasConcept C2777212361 @default.
- W3167827695 hasConcept C41008148 @default.
- W3167827695 hasConcept C41608201 @default.
- W3167827695 hasConcept C41895202 @default.
- W3167827695 hasConcept C57493831 @default.
- W3167827695 hasConcept C75294576 @default.
- W3167827695 hasConcept C97931131 @default.
- W3167827695 hasConceptScore W3167827695C103278499 @default.
- W3167827695 hasConceptScore W3167827695C111472728 @default.
- W3167827695 hasConceptScore W3167827695C11413529 @default.
- W3167827695 hasConceptScore W3167827695C115961682 @default.
- W3167827695 hasConceptScore W3167827695C119857082 @default.
- W3167827695 hasConceptScore W3167827695C124101348 @default.
- W3167827695 hasConceptScore W3167827695C124304363 @default.
- W3167827695 hasConceptScore W3167827695C138885662 @default.
- W3167827695 hasConceptScore W3167827695C153180895 @default.
- W3167827695 hasConceptScore W3167827695C154945302 @default.
- W3167827695 hasConceptScore W3167827695C162324750 @default.
- W3167827695 hasConceptScore W3167827695C176217482 @default.
- W3167827695 hasConceptScore W3167827695C177264268 @default.
- W3167827695 hasConceptScore W3167827695C199360897 @default.
- W3167827695 hasConceptScore W3167827695C21547014 @default.
- W3167827695 hasConceptScore W3167827695C2776401178 @default.
- W3167827695 hasConceptScore W3167827695C2777212361 @default.
- W3167827695 hasConceptScore W3167827695C41008148 @default.
- W3167827695 hasConceptScore W3167827695C41608201 @default.
- W3167827695 hasConceptScore W3167827695C41895202 @default.
- W3167827695 hasConceptScore W3167827695C57493831 @default.
- W3167827695 hasConceptScore W3167827695C75294576 @default.
- W3167827695 hasConceptScore W3167827695C97931131 @default.
- W3167827695 hasLocation W31678276951 @default.
- W3167827695 hasOpenAccess W3167827695 @default.
- W3167827695 hasPrimaryLocation W31678276951 @default.
- W3167827695 hasRelatedWork W2006186569 @default.
- W3167827695 hasRelatedWork W2755074597 @default.
- W3167827695 hasRelatedWork W2909196548 @default.
- W3167827695 hasRelatedWork W2923124246 @default.
- W3167827695 hasRelatedWork W2946155297 @default.
- W3167827695 hasRelatedWork W2952647681 @default.