Matches in SemOpenAlex for { <https://semopenalex.org/work/W4308236449> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4308236449 abstract "Multi-modal learning with both text and images benefits multiple applications, such as attribute extraction for e-commerce products. In this paper, we propose Cross-Modality Attention Contrastive Language-Image Pre-training (CMA-CLIP), a new multi-modal architecture to jointly learn the fine-grained inter-modality relationship. It fuses CLIP with a sequence-wise attention module and a modality-wise attention module. The network uses CLIP to bridge the inter-modality gap at the global level, and uses the sequence-wise attention module to capture the fine-grained alignment between text and images. Besides, it leverages a modality-wise attention module to learn the relevance of each modality to downstream tasks, making the network robust against irrelevant modalities. CMA-CLIP outperforms the state-of-the-art method on Fashion-Gen by 5.5% in accuracy, achieves competitive performance on Food101 and performance on par with the state-of-the-art method on MM-IMDb. We also demonstrate CMA-CLIP’s robustness against irrelevant modalities on an Amazon dataset for the task of product attribute extraction." @default.
- W4308236449 created "2022-11-09" @default.
- W4308236449 creator A5005218806 @default.
- W4308236449 creator A5013934000 @default.
- W4308236449 creator A5023363049 @default.
- W4308236449 creator A5046769911 @default.
- W4308236449 creator A5057117256 @default.
- W4308236449 creator A5058040690 @default.
- W4308236449 creator A5070138894 @default.
- W4308236449 creator A5071261572 @default.
- W4308236449 creator A5071941735 @default.
- W4308236449 date "2022-10-16" @default.
- W4308236449 modified "2023-10-16" @default.
- W4308236449 title "CMA-CLIP: Cross-Modality Attention Clip for Text-Image Classification" @default.
- W4308236449 cites W2886641317 @default.
- W4308236449 cites W3035485997 @default.
- W4308236449 cites W3165938948 @default.
- W4308236449 cites W3182683290 @default.
- W4308236449 cites W3213100861 @default.
- W4308236449 doi "https://doi.org/10.1109/icip46576.2022.9897323" @default.
- W4308236449 hasPublicationYear "2022" @default.
- W4308236449 type Work @default.
- W4308236449 citedByCount "3" @default.
- W4308236449 countsByYear W43082364492022 @default.
- W4308236449 countsByYear W43082364492023 @default.
- W4308236449 crossrefType "proceedings-article" @default.
- W4308236449 hasAuthorship W4308236449A5005218806 @default.
- W4308236449 hasAuthorship W4308236449A5013934000 @default.
- W4308236449 hasAuthorship W4308236449A5023363049 @default.
- W4308236449 hasAuthorship W4308236449A5046769911 @default.
- W4308236449 hasAuthorship W4308236449A5057117256 @default.
- W4308236449 hasAuthorship W4308236449A5058040690 @default.
- W4308236449 hasAuthorship W4308236449A5070138894 @default.
- W4308236449 hasAuthorship W4308236449A5071261572 @default.
- W4308236449 hasAuthorship W4308236449A5071941735 @default.
- W4308236449 hasConcept C104317684 @default.
- W4308236449 hasConcept C144024400 @default.
- W4308236449 hasConcept C153180895 @default.
- W4308236449 hasConcept C154945302 @default.
- W4308236449 hasConcept C158154518 @default.
- W4308236449 hasConcept C17744445 @default.
- W4308236449 hasConcept C185592680 @default.
- W4308236449 hasConcept C188027245 @default.
- W4308236449 hasConcept C199539241 @default.
- W4308236449 hasConcept C2779903281 @default.
- W4308236449 hasConcept C2780226545 @default.
- W4308236449 hasConcept C36289849 @default.
- W4308236449 hasConcept C41008148 @default.
- W4308236449 hasConcept C55493867 @default.
- W4308236449 hasConcept C63479239 @default.
- W4308236449 hasConcept C71139939 @default.
- W4308236449 hasConceptScore W4308236449C104317684 @default.
- W4308236449 hasConceptScore W4308236449C144024400 @default.
- W4308236449 hasConceptScore W4308236449C153180895 @default.
- W4308236449 hasConceptScore W4308236449C154945302 @default.
- W4308236449 hasConceptScore W4308236449C158154518 @default.
- W4308236449 hasConceptScore W4308236449C17744445 @default.
- W4308236449 hasConceptScore W4308236449C185592680 @default.
- W4308236449 hasConceptScore W4308236449C188027245 @default.
- W4308236449 hasConceptScore W4308236449C199539241 @default.
- W4308236449 hasConceptScore W4308236449C2779903281 @default.
- W4308236449 hasConceptScore W4308236449C2780226545 @default.
- W4308236449 hasConceptScore W4308236449C36289849 @default.
- W4308236449 hasConceptScore W4308236449C41008148 @default.
- W4308236449 hasConceptScore W4308236449C55493867 @default.
- W4308236449 hasConceptScore W4308236449C63479239 @default.
- W4308236449 hasConceptScore W4308236449C71139939 @default.
- W4308236449 hasLocation W43082364491 @default.
- W4308236449 hasOpenAccess W4308236449 @default.
- W4308236449 hasPrimaryLocation W43082364491 @default.
- W4308236449 hasRelatedWork W2474574787 @default.
- W4308236449 hasRelatedWork W2949074159 @default.
- W4308236449 hasRelatedWork W2952745240 @default.
- W4308236449 hasRelatedWork W3142456083 @default.
- W4308236449 hasRelatedWork W3198037411 @default.
- W4308236449 hasRelatedWork W4226120250 @default.
- W4308236449 hasRelatedWork W4298715519 @default.
- W4308236449 hasRelatedWork W4301143707 @default.
- W4308236449 hasRelatedWork W4306353150 @default.
- W4308236449 hasRelatedWork W4386721968 @default.
- W4308236449 isParatext "false" @default.
- W4308236449 isRetracted "false" @default.
- W4308236449 workType "article" @default.