Matches in SemOpenAlex for { <https://semopenalex.org/work/W4225547495> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W4225547495 abstract "Multi-label few-shot image classification (ML-FSIC) is the task of assigning descriptive labels to previously unseen images, based on a small number of training examples. A key feature of the multi-label setting is that images often have multiple labels, which typically refer to different regions of the image. When estimating prototypes, in a metric-based setting, it is thus important to determine which regions are relevant for which labels, but the limited amount of training data makes this highly challenging. As a solution, in this paper we propose to use word embeddings as a form of prior knowledge about the meaning of the labels. In particular, visual prototypes are obtained by aggregating the local feature maps of the support images, using an attention mechanism that relies on the label embeddings. As an important advantage, our model can infer prototypes for unseen labels without the need for fine-tuning any model parameters, which demonstrates its strong generalization abilities. Experiments on COCO and PASCAL VOC furthermore show that our model substantially improves the current state-of-the-art." @default.
- W4225547495 created "2022-05-05" @default.
- W4225547495 creator A5007754802 @default.
- W4225547495 creator A5009887598 @default.
- W4225547495 creator A5013736647 @default.
- W4225547495 creator A5018109582 @default.
- W4225547495 creator A5059712495 @default.
- W4225547495 creator A5059764145 @default.
- W4225547495 creator A5082611298 @default.
- W4225547495 date "2021-12-02" @default.
- W4225547495 modified "2023-10-14" @default.
- W4225547495 title "Inferring Prototypes for Multi-Label Few-Shot Image Classification with Word Vector Guided Attention" @default.
- W4225547495 hasPublicationYear "2021" @default.
- W4225547495 type Work @default.
- W4225547495 citedByCount "0" @default.
- W4225547495 crossrefType "posted-content" @default.
- W4225547495 hasAuthorship W4225547495A5007754802 @default.
- W4225547495 hasAuthorship W4225547495A5009887598 @default.
- W4225547495 hasAuthorship W4225547495A5013736647 @default.
- W4225547495 hasAuthorship W4225547495A5018109582 @default.
- W4225547495 hasAuthorship W4225547495A5059712495 @default.
- W4225547495 hasAuthorship W4225547495A5059764145 @default.
- W4225547495 hasAuthorship W4225547495A5082611298 @default.
- W4225547495 hasBestOaLocation W42255474951 @default.
- W4225547495 hasConcept C115961682 @default.
- W4225547495 hasConcept C119857082 @default.
- W4225547495 hasConcept C134306372 @default.
- W4225547495 hasConcept C138885662 @default.
- W4225547495 hasConcept C153180895 @default.
- W4225547495 hasConcept C154945302 @default.
- W4225547495 hasConcept C162324750 @default.
- W4225547495 hasConcept C176217482 @default.
- W4225547495 hasConcept C177148314 @default.
- W4225547495 hasConcept C199360897 @default.
- W4225547495 hasConcept C204321447 @default.
- W4225547495 hasConcept C21547014 @default.
- W4225547495 hasConcept C2524010 @default.
- W4225547495 hasConcept C2776401178 @default.
- W4225547495 hasConcept C33923547 @default.
- W4225547495 hasConcept C41008148 @default.
- W4225547495 hasConcept C41895202 @default.
- W4225547495 hasConcept C75608658 @default.
- W4225547495 hasConcept C83665646 @default.
- W4225547495 hasConcept C90805587 @default.
- W4225547495 hasConceptScore W4225547495C115961682 @default.
- W4225547495 hasConceptScore W4225547495C119857082 @default.
- W4225547495 hasConceptScore W4225547495C134306372 @default.
- W4225547495 hasConceptScore W4225547495C138885662 @default.
- W4225547495 hasConceptScore W4225547495C153180895 @default.
- W4225547495 hasConceptScore W4225547495C154945302 @default.
- W4225547495 hasConceptScore W4225547495C162324750 @default.
- W4225547495 hasConceptScore W4225547495C176217482 @default.
- W4225547495 hasConceptScore W4225547495C177148314 @default.
- W4225547495 hasConceptScore W4225547495C199360897 @default.
- W4225547495 hasConceptScore W4225547495C204321447 @default.
- W4225547495 hasConceptScore W4225547495C21547014 @default.
- W4225547495 hasConceptScore W4225547495C2524010 @default.
- W4225547495 hasConceptScore W4225547495C2776401178 @default.
- W4225547495 hasConceptScore W4225547495C33923547 @default.
- W4225547495 hasConceptScore W4225547495C41008148 @default.
- W4225547495 hasConceptScore W4225547495C41895202 @default.
- W4225547495 hasConceptScore W4225547495C75608658 @default.
- W4225547495 hasConceptScore W4225547495C83665646 @default.
- W4225547495 hasConceptScore W4225547495C90805587 @default.
- W4225547495 hasLocation W42255474951 @default.
- W4225547495 hasOpenAccess W4225547495 @default.
- W4225547495 hasPrimaryLocation W42255474951 @default.
- W4225547495 hasRelatedWork W1000462 @default.
- W4225547495 hasRelatedWork W12335020 @default.
- W4225547495 hasRelatedWork W12820539 @default.
- W4225547495 hasRelatedWork W13426584 @default.
- W4225547495 hasRelatedWork W1383942 @default.
- W4225547495 hasRelatedWork W1496815 @default.
- W4225547495 hasRelatedWork W15119441 @default.
- W4225547495 hasRelatedWork W16066671 @default.
- W4225547495 hasRelatedWork W701070 @default.
- W4225547495 hasRelatedWork W9958333 @default.
- W4225547495 isParatext "false" @default.
- W4225547495 isRetracted "false" @default.
- W4225547495 workType "article" @default.