Matches in SemOpenAlex for { <https://semopenalex.org/work/W3208696560> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W3208696560 abstract "This paper presents a three-tier modality alignment approach to learning text-image joint embedding, coined as JEMA, for cross-modal retrieval of cooking recipes and food images. The first tier improves recipe text embedding by optimizing the LSTM networks with term extraction and ranking enhanced sequence patterns, and optimizes the image embedding by combining the ResNeXt-101 image encoder with the category embedding using wideResNet-50 with word2vec. The second tier modality alignment optimizes the textual-visual joint embedding loss function using a double batch-hard triplet loss with soft-margin optimization. The third modality alignment incorporates two types of cross-modality alignments as the auxiliary loss regularizations to further reduce the alignment errors in the joint learning of the two modality-specific embedding functions. The category-based cross-modal alignment aims to align the image category with the recipe category as a loss regularization to the joint embedding. The cross-modal discriminator-based alignment aims to add the visual-textual embedding distribution alignment to further regularize the joint embedding loss. Extensive experiments with the one-million recipes benchmark dataset Recipe1M demonstrate that the proposed JEMA approach outperforms the state-of-the-art cross-modal embedding methods for both image-to-recipe and recipe-to-image retrievals." @default.
- W3208696560 created "2021-11-08" @default.
- W3208696560 creator A5003285069 @default.
- W3208696560 creator A5037897744 @default.
- W3208696560 creator A5037906509 @default.
- W3208696560 creator A5038310950 @default.
- W3208696560 date "2021-10-26" @default.
- W3208696560 modified "2023-09-24" @default.
- W3208696560 title "Learning Joint Embedding with Modality Alignments for Cross-Modal Retrieval of Recipes and Food Images" @default.
- W3208696560 cites W12634471 @default.
- W3208696560 cites W1964073652 @default.
- W3208696560 cites W1978394996 @default.
- W3208696560 cites W1982635469 @default.
- W3208696560 cites W2092361219 @default.
- W3208696560 cites W2106277773 @default.
- W3208696560 cites W2137918516 @default.
- W3208696560 cites W2194775991 @default.
- W3208696560 cites W2293499654 @default.
- W3208696560 cites W2526198870 @default.
- W3208696560 cites W2549139847 @default.
- W3208696560 cites W2565041183 @default.
- W3208696560 cites W2604325789 @default.
- W3208696560 cites W2728515412 @default.
- W3208696560 cites W2737041163 @default.
- W3208696560 cites W2765440071 @default.
- W3208696560 cites W2767221212 @default.
- W3208696560 cites W2897152025 @default.
- W3208696560 cites W2948037078 @default.
- W3208696560 cites W2963055199 @default.
- W3208696560 cites W2963997278 @default.
- W3208696560 cites W3034303366 @default.
- W3208696560 cites W3035032757 @default.
- W3208696560 doi "https://doi.org/10.1145/3459637.3482270" @default.
- W3208696560 hasPublicationYear "2021" @default.
- W3208696560 type Work @default.
- W3208696560 sameAs 3208696560 @default.
- W3208696560 citedByCount "5" @default.
- W3208696560 countsByYear W32086965602022 @default.
- W3208696560 countsByYear W32086965602023 @default.
- W3208696560 crossrefType "proceedings-article" @default.
- W3208696560 hasAuthorship W3208696560A5003285069 @default.
- W3208696560 hasAuthorship W3208696560A5037897744 @default.
- W3208696560 hasAuthorship W3208696560A5037906509 @default.
- W3208696560 hasAuthorship W3208696560A5038310950 @default.
- W3208696560 hasBestOaLocation W32086965602 @default.
- W3208696560 hasConcept C103278499 @default.
- W3208696560 hasConcept C115961682 @default.
- W3208696560 hasConcept C13280743 @default.
- W3208696560 hasConcept C153180895 @default.
- W3208696560 hasConcept C154945302 @default.
- W3208696560 hasConcept C185592680 @default.
- W3208696560 hasConcept C185798385 @default.
- W3208696560 hasConcept C188027245 @default.
- W3208696560 hasConcept C205649164 @default.
- W3208696560 hasConcept C2780226545 @default.
- W3208696560 hasConcept C31972630 @default.
- W3208696560 hasConcept C41008148 @default.
- W3208696560 hasConcept C41608201 @default.
- W3208696560 hasConcept C71139939 @default.
- W3208696560 hasConceptScore W3208696560C103278499 @default.
- W3208696560 hasConceptScore W3208696560C115961682 @default.
- W3208696560 hasConceptScore W3208696560C13280743 @default.
- W3208696560 hasConceptScore W3208696560C153180895 @default.
- W3208696560 hasConceptScore W3208696560C154945302 @default.
- W3208696560 hasConceptScore W3208696560C185592680 @default.
- W3208696560 hasConceptScore W3208696560C185798385 @default.
- W3208696560 hasConceptScore W3208696560C188027245 @default.
- W3208696560 hasConceptScore W3208696560C205649164 @default.
- W3208696560 hasConceptScore W3208696560C2780226545 @default.
- W3208696560 hasConceptScore W3208696560C31972630 @default.
- W3208696560 hasConceptScore W3208696560C41008148 @default.
- W3208696560 hasConceptScore W3208696560C41608201 @default.
- W3208696560 hasConceptScore W3208696560C71139939 @default.
- W3208696560 hasLocation W32086965601 @default.
- W3208696560 hasLocation W32086965602 @default.
- W3208696560 hasOpenAccess W3208696560 @default.
- W3208696560 hasPrimaryLocation W32086965601 @default.
- W3208696560 hasRelatedWork W2005185696 @default.
- W3208696560 hasRelatedWork W2015538044 @default.
- W3208696560 hasRelatedWork W2130228941 @default.
- W3208696560 hasRelatedWork W2161229648 @default.
- W3208696560 hasRelatedWork W2164688428 @default.
- W3208696560 hasRelatedWork W2235753890 @default.
- W3208696560 hasRelatedWork W2798513620 @default.
- W3208696560 hasRelatedWork W2993674027 @default.
- W3208696560 hasRelatedWork W3208409104 @default.
- W3208696560 hasRelatedWork W4299906651 @default.
- W3208696560 isParatext "false" @default.
- W3208696560 isRetracted "false" @default.
- W3208696560 magId "3208696560" @default.
- W3208696560 workType "article" @default.