Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385894680> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W4385894680 abstract "Multi-label image recognition in the low-label regime is a task of great challenge and practical significance. Previous works have focused on learning the alignment between textual and visual spaces to compensate for limited image labels, yet may suffer from reduced accuracy due to the scarcity of high-quality multi-label annotations. In this research, we leverage the powerful alignment between textual and visual features pretrained with millions of auxiliary image-text pairs. We introduce an efficient and effective framework called Evidence-guided Dual Context Optimization (DualCoOp++), which serves as a unified approach for addressing partial-label and zero-shot multi-label recognition. In DualCoOp++ we separately encode evidential, positive, and negative contexts for target classes as parametric components of the linguistic input (i.e., prompts). The evidential context aims to discover all the related visual content for the target class, and serves as guidance to aggregate positive and negative contexts from the spatial domain of the image, enabling better distinguishment between similar categories. Additionally, we introduce a Winner-Take-All module that promotes inter-class interaction during training, while avoiding the need for extra parameters and costs. As DualCoOp++ imposes minimal additional learnable overhead on the pretrained vision-language framework, it enables rapid adaptation to multi-label recognition tasks with limited annotations and even unseen classes. Experiments on standard multi-label recognition benchmarks across two challenging low-label settings demonstrate the superior performance of our approach compared to state-of-the-art methods." @default.
- W4385894680 created "2023-08-18" @default.
- W4385894680 creator A5021031074 @default.
- W4385894680 creator A5042663603 @default.
- W4385894680 creator A5066289448 @default.
- W4385894680 creator A5075906727 @default.
- W4385894680 date "2023-08-03" @default.
- W4385894680 modified "2023-09-23" @default.
- W4385894680 title "DualCoOp++: Fast and Effective Adaptation to Multi-Label Recognition with Limited Annotations" @default.
- W4385894680 doi "https://doi.org/10.48550/arxiv.2308.01890" @default.
- W4385894680 hasPublicationYear "2023" @default.
- W4385894680 type Work @default.
- W4385894680 citedByCount "0" @default.
- W4385894680 crossrefType "posted-content" @default.
- W4385894680 hasAuthorship W4385894680A5021031074 @default.
- W4385894680 hasAuthorship W4385894680A5042663603 @default.
- W4385894680 hasAuthorship W4385894680A5066289448 @default.
- W4385894680 hasAuthorship W4385894680A5075906727 @default.
- W4385894680 hasBestOaLocation W43858946801 @default.
- W4385894680 hasConcept C119857082 @default.
- W4385894680 hasConcept C120665830 @default.
- W4385894680 hasConcept C121332964 @default.
- W4385894680 hasConcept C139807058 @default.
- W4385894680 hasConcept C151730666 @default.
- W4385894680 hasConcept C153083717 @default.
- W4385894680 hasConcept C153180895 @default.
- W4385894680 hasConcept C154945302 @default.
- W4385894680 hasConcept C162324750 @default.
- W4385894680 hasConcept C187736073 @default.
- W4385894680 hasConcept C204321447 @default.
- W4385894680 hasConcept C2777212361 @default.
- W4385894680 hasConcept C2779343474 @default.
- W4385894680 hasConcept C2780451532 @default.
- W4385894680 hasConcept C41008148 @default.
- W4385894680 hasConcept C86803240 @default.
- W4385894680 hasConceptScore W4385894680C119857082 @default.
- W4385894680 hasConceptScore W4385894680C120665830 @default.
- W4385894680 hasConceptScore W4385894680C121332964 @default.
- W4385894680 hasConceptScore W4385894680C139807058 @default.
- W4385894680 hasConceptScore W4385894680C151730666 @default.
- W4385894680 hasConceptScore W4385894680C153083717 @default.
- W4385894680 hasConceptScore W4385894680C153180895 @default.
- W4385894680 hasConceptScore W4385894680C154945302 @default.
- W4385894680 hasConceptScore W4385894680C162324750 @default.
- W4385894680 hasConceptScore W4385894680C187736073 @default.
- W4385894680 hasConceptScore W4385894680C204321447 @default.
- W4385894680 hasConceptScore W4385894680C2777212361 @default.
- W4385894680 hasConceptScore W4385894680C2779343474 @default.
- W4385894680 hasConceptScore W4385894680C2780451532 @default.
- W4385894680 hasConceptScore W4385894680C41008148 @default.
- W4385894680 hasConceptScore W4385894680C86803240 @default.
- W4385894680 hasLocation W43858946801 @default.
- W4385894680 hasOpenAccess W4385894680 @default.
- W4385894680 hasPrimaryLocation W43858946801 @default.
- W4385894680 hasRelatedWork W1571518467 @default.
- W4385894680 hasRelatedWork W2081647779 @default.
- W4385894680 hasRelatedWork W2961085424 @default.
- W4385894680 hasRelatedWork W3046775127 @default.
- W4385894680 hasRelatedWork W4285260836 @default.
- W4385894680 hasRelatedWork W4286629047 @default.
- W4385894680 hasRelatedWork W4306321456 @default.
- W4385894680 hasRelatedWork W4306674287 @default.
- W4385894680 hasRelatedWork W87991986 @default.
- W4385894680 hasRelatedWork W4224009465 @default.
- W4385894680 isParatext "false" @default.
- W4385894680 isRetracted "false" @default.
- W4385894680 workType "article" @default.