Matches in SemOpenAlex for { <https://semopenalex.org/work/W4285221285> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W4285221285 endingPage "1" @default.
- W4285221285 startingPage "1" @default.
- W4285221285 abstract "Cross-modal retrieval aims at retrieving highly semantic relevant information among multi-modalities. Existing cross-modal retrieval methods mainly explore the semantic consistency between image and text while rarely consider the rankings of positive instances in the retrieval results. Moreover, these methods seldom take into account the cross-interaction between image and text, which leads to the deficiency of learning their semantic relations. In this paper, we propose a Unified framework with Ranking Learning (URL) for cross-modal retrieval. The unified framework consists of three sub-networks, visual network, textual network, and interaction network. Visual network and textual network project the image feature and text feature into their corresponding hidden spaces respectively. Then, the interaction network forces the target image-text representation to align in the common space. For unifying both semantics and rankings, we propose a new optimization paradigm including pre-alignment for semantic knowledge transfer and ranking learning for final retrieval, which can decouple semantic alignment and ranking learning. The former focuses on the semantic pre-alignment optimized by semantic classification and the latter revolves around the retrieval rankings. For the ranking learning, we introduce a cross-AP loss which can directly optimize the retrieval metric average precision for cross-modal retrieval. We conduct experiments on four widely-used benchmarks, including Wikipedia dataset, Pascal Sentence dataset, NUS-WIDE-10k dataset, and PKU XMediaNet dataset respectively. Extensive experimental results show that the proposed method can obtain higher retrieval precision." @default.
- W4285221285 created "2022-07-14" @default.
- W4285221285 creator A5000141628 @default.
- W4285221285 creator A5004061050 @default.
- W4285221285 creator A5021486674 @default.
- W4285221285 creator A5064548129 @default.
- W4285221285 creator A5064787764 @default.
- W4285221285 date "2022-01-01" @default.
- W4285221285 modified "2023-10-14" @default.
- W4285221285 title "Semantic Pre-alignment and Ranking Learning with Unified Framework for Cross-modal Retrieval" @default.
- W4285221285 doi "https://doi.org/10.1109/tcsvt.2022.3182549" @default.
- W4285221285 hasPublicationYear "2022" @default.
- W4285221285 type Work @default.
- W4285221285 citedByCount "10" @default.
- W4285221285 countsByYear W42852212852022 @default.
- W4285221285 countsByYear W42852212852023 @default.
- W4285221285 crossrefType "journal-article" @default.
- W4285221285 hasAuthorship W4285221285A5000141628 @default.
- W4285221285 hasAuthorship W4285221285A5004061050 @default.
- W4285221285 hasAuthorship W4285221285A5021486674 @default.
- W4285221285 hasAuthorship W4285221285A5064548129 @default.
- W4285221285 hasAuthorship W4285221285A5064787764 @default.
- W4285221285 hasConcept C115961682 @default.
- W4285221285 hasConcept C138885662 @default.
- W4285221285 hasConcept C154945302 @default.
- W4285221285 hasConcept C1667742 @default.
- W4285221285 hasConcept C173862523 @default.
- W4285221285 hasConcept C184337299 @default.
- W4285221285 hasConcept C189391414 @default.
- W4285221285 hasConcept C189430467 @default.
- W4285221285 hasConcept C199360897 @default.
- W4285221285 hasConcept C2129575 @default.
- W4285221285 hasConcept C23123220 @default.
- W4285221285 hasConcept C2776401178 @default.
- W4285221285 hasConcept C2776436953 @default.
- W4285221285 hasConcept C41008148 @default.
- W4285221285 hasConcept C41895202 @default.
- W4285221285 hasConcept C511149849 @default.
- W4285221285 hasConcept C6881194 @default.
- W4285221285 hasConcept C86037889 @default.
- W4285221285 hasConceptScore W4285221285C115961682 @default.
- W4285221285 hasConceptScore W4285221285C138885662 @default.
- W4285221285 hasConceptScore W4285221285C154945302 @default.
- W4285221285 hasConceptScore W4285221285C1667742 @default.
- W4285221285 hasConceptScore W4285221285C173862523 @default.
- W4285221285 hasConceptScore W4285221285C184337299 @default.
- W4285221285 hasConceptScore W4285221285C189391414 @default.
- W4285221285 hasConceptScore W4285221285C189430467 @default.
- W4285221285 hasConceptScore W4285221285C199360897 @default.
- W4285221285 hasConceptScore W4285221285C2129575 @default.
- W4285221285 hasConceptScore W4285221285C23123220 @default.
- W4285221285 hasConceptScore W4285221285C2776401178 @default.
- W4285221285 hasConceptScore W4285221285C2776436953 @default.
- W4285221285 hasConceptScore W4285221285C41008148 @default.
- W4285221285 hasConceptScore W4285221285C41895202 @default.
- W4285221285 hasConceptScore W4285221285C511149849 @default.
- W4285221285 hasConceptScore W4285221285C6881194 @default.
- W4285221285 hasConceptScore W4285221285C86037889 @default.
- W4285221285 hasFunder F4320321001 @default.
- W4285221285 hasLocation W42852212851 @default.
- W4285221285 hasOpenAccess W4285221285 @default.
- W4285221285 hasPrimaryLocation W42852212851 @default.
- W4285221285 hasRelatedWork W1974970223 @default.
- W4285221285 hasRelatedWork W2031812225 @default.
- W4285221285 hasRelatedWork W2088097596 @default.
- W4285221285 hasRelatedWork W2113661533 @default.
- W4285221285 hasRelatedWork W2151414079 @default.
- W4285221285 hasRelatedWork W2366482673 @default.
- W4285221285 hasRelatedWork W2540793605 @default.
- W4285221285 hasRelatedWork W2545852610 @default.
- W4285221285 hasRelatedWork W2548806402 @default.
- W4285221285 hasRelatedWork W64345524 @default.
- W4285221285 isParatext "false" @default.
- W4285221285 isRetracted "false" @default.
- W4285221285 workType "article" @default.