Matches in SemOpenAlex for { <https://semopenalex.org/work/W4327810948> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W4327810948 abstract "Recent Few-Shot Learning (FSL) methods put emphasis on generating a discriminative embedding features to precisely measure the similarity between support and query sets. Current CNN-based cross-attention approaches generate discriminative representations via enhancing the mutually semantic similar regions of support and query pairs. However, it suffers from two problems: CNN structure produces inaccurate attention map based on local features, and mutually similar backgrounds cause distraction. To alleviate these problems, we design a novel SpatialFormer structure to generate more accurate attention regions based on global features. Different from the traditional Transformer modeling intrinsic instance-level similarity which causes accuracy degradation in FSL, our SpatialFormer explores the semantic-level similarity between pair inputs to boost the performance. Then we derive two specific attention modules, named SpatialFormer Semantic Attention (SFSA) and SpatialFormer Target Attention (SFTA), to enhance the target object regions while reduce the background distraction. Particularly, SFSA highlights the regions with same semantic information between pair features, and SFTA finds potential foreground object regions of novel feature that are similar to base categories. Extensive experiments show that our methods are effective and achieve new state-of-the-art results on few-shot classification benchmarks." @default.
- W4327810948 created "2023-03-19" @default.
- W4327810948 creator A5000290451 @default.
- W4327810948 creator A5017665110 @default.
- W4327810948 creator A5020568230 @default.
- W4327810948 creator A5023834700 @default.
- W4327810948 creator A5027835055 @default.
- W4327810948 creator A5032912558 @default.
- W4327810948 creator A5042532963 @default.
- W4327810948 creator A5043731569 @default.
- W4327810948 creator A5057352675 @default.
- W4327810948 creator A5076699095 @default.
- W4327810948 creator A5077068949 @default.
- W4327810948 date "2023-03-15" @default.
- W4327810948 modified "2023-10-01" @default.
- W4327810948 title "SpatialFormer: Semantic and Target Aware Attentions for Few-Shot Learning" @default.
- W4327810948 doi "https://doi.org/10.48550/arxiv.2303.09281" @default.
- W4327810948 hasPublicationYear "2023" @default.
- W4327810948 type Work @default.
- W4327810948 citedByCount "0" @default.
- W4327810948 crossrefType "posted-content" @default.
- W4327810948 hasAuthorship W4327810948A5000290451 @default.
- W4327810948 hasAuthorship W4327810948A5017665110 @default.
- W4327810948 hasAuthorship W4327810948A5020568230 @default.
- W4327810948 hasAuthorship W4327810948A5023834700 @default.
- W4327810948 hasAuthorship W4327810948A5027835055 @default.
- W4327810948 hasAuthorship W4327810948A5032912558 @default.
- W4327810948 hasAuthorship W4327810948A5042532963 @default.
- W4327810948 hasAuthorship W4327810948A5043731569 @default.
- W4327810948 hasAuthorship W4327810948A5057352675 @default.
- W4327810948 hasAuthorship W4327810948A5076699095 @default.
- W4327810948 hasAuthorship W4327810948A5077068949 @default.
- W4327810948 hasBestOaLocation W43278109481 @default.
- W4327810948 hasConcept C103278499 @default.
- W4327810948 hasConcept C115961682 @default.
- W4327810948 hasConcept C119857082 @default.
- W4327810948 hasConcept C121332964 @default.
- W4327810948 hasConcept C130318100 @default.
- W4327810948 hasConcept C138885662 @default.
- W4327810948 hasConcept C153180895 @default.
- W4327810948 hasConcept C154945302 @default.
- W4327810948 hasConcept C165801399 @default.
- W4327810948 hasConcept C169760540 @default.
- W4327810948 hasConcept C204321447 @default.
- W4327810948 hasConcept C2776378700 @default.
- W4327810948 hasConcept C2776401178 @default.
- W4327810948 hasConcept C2781238097 @default.
- W4327810948 hasConcept C41008148 @default.
- W4327810948 hasConcept C41608201 @default.
- W4327810948 hasConcept C41895202 @default.
- W4327810948 hasConcept C62520636 @default.
- W4327810948 hasConcept C66322947 @default.
- W4327810948 hasConcept C86803240 @default.
- W4327810948 hasConcept C97931131 @default.
- W4327810948 hasConceptScore W4327810948C103278499 @default.
- W4327810948 hasConceptScore W4327810948C115961682 @default.
- W4327810948 hasConceptScore W4327810948C119857082 @default.
- W4327810948 hasConceptScore W4327810948C121332964 @default.
- W4327810948 hasConceptScore W4327810948C130318100 @default.
- W4327810948 hasConceptScore W4327810948C138885662 @default.
- W4327810948 hasConceptScore W4327810948C153180895 @default.
- W4327810948 hasConceptScore W4327810948C154945302 @default.
- W4327810948 hasConceptScore W4327810948C165801399 @default.
- W4327810948 hasConceptScore W4327810948C169760540 @default.
- W4327810948 hasConceptScore W4327810948C204321447 @default.
- W4327810948 hasConceptScore W4327810948C2776378700 @default.
- W4327810948 hasConceptScore W4327810948C2776401178 @default.
- W4327810948 hasConceptScore W4327810948C2781238097 @default.
- W4327810948 hasConceptScore W4327810948C41008148 @default.
- W4327810948 hasConceptScore W4327810948C41608201 @default.
- W4327810948 hasConceptScore W4327810948C41895202 @default.
- W4327810948 hasConceptScore W4327810948C62520636 @default.
- W4327810948 hasConceptScore W4327810948C66322947 @default.
- W4327810948 hasConceptScore W4327810948C86803240 @default.
- W4327810948 hasConceptScore W4327810948C97931131 @default.
- W4327810948 hasLocation W43278109481 @default.
- W4327810948 hasOpenAccess W4327810948 @default.
- W4327810948 hasPrimaryLocation W43278109481 @default.
- W4327810948 hasRelatedWork W1652783584 @default.
- W4327810948 hasRelatedWork W1971623867 @default.
- W4327810948 hasRelatedWork W1982770690 @default.
- W4327810948 hasRelatedWork W2015538044 @default.
- W4327810948 hasRelatedWork W2024160000 @default.
- W4327810948 hasRelatedWork W2038246283 @default.
- W4327810948 hasRelatedWork W2404514746 @default.
- W4327810948 hasRelatedWork W3202305871 @default.
- W4327810948 hasRelatedWork W4281690070 @default.
- W4327810948 hasRelatedWork W4289653936 @default.
- W4327810948 isParatext "false" @default.
- W4327810948 isRetracted "false" @default.
- W4327810948 workType "article" @default.