Matches in SemOpenAlex for { <https://semopenalex.org/work/W3034291519> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W3034291519 abstract "Visual dialog is a challenging task that requires the comprehension of the semantic dependencies among implicit visual and textual contexts. This task can refer to the relation inference in a graphical model with sparse contexts and unknown graph structure (relation descriptor), and how to model the underlying context-aware relation inference is critical. To this end, we propose a novel Context-Aware Graph (CAG) neural network. Each node in the graph corresponds to a joint semantic feature, including both object-based (visual) and history-related (textual) context representations. The graph structure (relations in dialog) is iteratively updated using an adaptive top-K message passing mechanism. Specifically, in every message passing step, each node selects the most K relevant nodes, and only receives messages from them. Then, after the update, we impose graph attention on all the nodes to get the final graph embedding and infer the answer. In CAG, each node has dynamic relations in the graph (different related K neighbor nodes), and only the most relevant nodes are attributive to the context-aware relational graph inference. Experimental results on VisDial v0.9 and v1.0 datasets show that CAG outperforms comparative methods. Visualization results further validate the interpretability of our method." @default.
- W3034291519 created "2020-06-19" @default.
- W3034291519 creator A5003217535 @default.
- W3034291519 creator A5014753115 @default.
- W3034291519 creator A5036726873 @default.
- W3034291519 creator A5042324027 @default.
- W3034291519 creator A5059530979 @default.
- W3034291519 date "2020-06-01" @default.
- W3034291519 modified "2023-09-30" @default.
- W3034291519 title "Iterative Context-Aware Graph Inference for Visual Dialog" @default.
- W3034291519 cites W2250539671 @default.
- W3034291519 cites W2550553598 @default.
- W3034291519 cites W2745461083 @default.
- W3034291519 cites W2798990097 @default.
- W3034291519 cites W2799215407 @default.
- W3034291519 cites W2887585070 @default.
- W3034291519 cites W2890718122 @default.
- W3034291519 cites W2904910963 @default.
- W3034291519 cites W2917061951 @default.
- W3034291519 cites W2938603906 @default.
- W3034291519 cites W2940457086 @default.
- W3034291519 cites W2950096400 @default.
- W3034291519 cites W2953016680 @default.
- W3034291519 cites W2962779575 @default.
- W3034291519 cites W2963287297 @default.
- W3034291519 cites W2963623904 @default.
- W3034291519 cites W2963735856 @default.
- W3034291519 cites W2964057271 @default.
- W3034291519 cites W2964118342 @default.
- W3034291519 cites W2964218959 @default.
- W3034291519 cites W2966158321 @default.
- W3034291519 cites W2966683369 @default.
- W3034291519 cites W2967045987 @default.
- W3034291519 cites W2970355596 @default.
- W3034291519 cites W2981902456 @default.
- W3034291519 cites W2987401211 @default.
- W3034291519 cites W2987671777 @default.
- W3034291519 cites W2988023442 @default.
- W3034291519 doi "https://doi.org/10.1109/cvpr42600.2020.01007" @default.
- W3034291519 hasPublicationYear "2020" @default.
- W3034291519 type Work @default.
- W3034291519 sameAs 3034291519 @default.
- W3034291519 citedByCount "35" @default.
- W3034291519 countsByYear W30342915192020 @default.
- W3034291519 countsByYear W30342915192021 @default.
- W3034291519 countsByYear W30342915192022 @default.
- W3034291519 countsByYear W30342915192023 @default.
- W3034291519 crossrefType "proceedings-article" @default.
- W3034291519 hasAuthorship W3034291519A5003217535 @default.
- W3034291519 hasAuthorship W3034291519A5014753115 @default.
- W3034291519 hasAuthorship W3034291519A5036726873 @default.
- W3034291519 hasAuthorship W3034291519A5042324027 @default.
- W3034291519 hasAuthorship W3034291519A5059530979 @default.
- W3034291519 hasBestOaLocation W30342915192 @default.
- W3034291519 hasConcept C132525143 @default.
- W3034291519 hasConcept C136764020 @default.
- W3034291519 hasConcept C154945302 @default.
- W3034291519 hasConcept C173853756 @default.
- W3034291519 hasConcept C204321447 @default.
- W3034291519 hasConcept C2776214188 @default.
- W3034291519 hasConcept C2781067378 @default.
- W3034291519 hasConcept C41008148 @default.
- W3034291519 hasConcept C80444323 @default.
- W3034291519 hasConceptScore W3034291519C132525143 @default.
- W3034291519 hasConceptScore W3034291519C136764020 @default.
- W3034291519 hasConceptScore W3034291519C154945302 @default.
- W3034291519 hasConceptScore W3034291519C173853756 @default.
- W3034291519 hasConceptScore W3034291519C204321447 @default.
- W3034291519 hasConceptScore W3034291519C2776214188 @default.
- W3034291519 hasConceptScore W3034291519C2781067378 @default.
- W3034291519 hasConceptScore W3034291519C41008148 @default.
- W3034291519 hasConceptScore W3034291519C80444323 @default.
- W3034291519 hasLocation W30342915191 @default.
- W3034291519 hasLocation W30342915192 @default.
- W3034291519 hasOpenAccess W3034291519 @default.
- W3034291519 hasPrimaryLocation W30342915191 @default.
- W3034291519 hasRelatedWork W1862650538 @default.
- W3034291519 hasRelatedWork W2352008582 @default.
- W3034291519 hasRelatedWork W2355108509 @default.
- W3034291519 hasRelatedWork W2898884225 @default.
- W3034291519 hasRelatedWork W2935694273 @default.
- W3034291519 hasRelatedWork W2964218959 @default.
- W3034291519 hasRelatedWork W2968260065 @default.
- W3034291519 hasRelatedWork W4206534706 @default.
- W3034291519 hasRelatedWork W4289306494 @default.
- W3034291519 hasRelatedWork W1872130062 @default.
- W3034291519 isParatext "false" @default.
- W3034291519 isRetracted "false" @default.
- W3034291519 magId "3034291519" @default.
- W3034291519 workType "article" @default.