Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386072088> ?p ?o ?g. }
- W4386072088 abstract "In this paper, we extend scene understanding to include that of human sketch. The result is a complete trilogy of scene representation from three diverse and complementary modalities – sketch, photo, and text. Instead of learning a rigid three-way embedding and be done with it, wefocus on learning a flexible joint embedding that fully supports the “optionality” that this complementarity brings. Our embedding supports optionality on two axes: (i) optionality across modalities – use any combination of modalities as query for downstream tasks like retrieval, (ii) optionality across tasks – simultaneously utilising the embedding for either discriminative (e.g., retrieval) or generative tasks (e.g., captioning). This provides flexibility to end-users by exploiting the best of each modality, therefore serving the very purpose behind our proposal of a trilogy in the first place. First, a combination of information-bottleneck and conditional invertible neural networks disentangle the modality-specific component from modality-agnostic in sketch, photo, and text. Second, the modality-agnostic instances from sketch, photo, and text are synergised using a modified cross-attention. Once learned, we show our embedding can accommodate a multi-facet of scene-related tasks, including those enabled for the first time by the inclusion of sketch, all without any task-specific modifications. Project Page: https://pinakinathc.github.io/scenetrilogy" @default.
- W4386072088 created "2023-08-23" @default.
- W4386072088 creator A5003628834 @default.
- W4386072088 creator A5010062829 @default.
- W4386072088 creator A5014436524 @default.
- W4386072088 creator A5043791403 @default.
- W4386072088 creator A5046046128 @default.
- W4386072088 creator A5085178555 @default.
- W4386072088 date "2023-06-01" @default.
- W4386072088 modified "2023-10-16" @default.
- W4386072088 title "SceneTrilogy: On Human Scene-Sketch and its Complementarity with Photo and Text" @default.
- W4386072088 cites W1905882502 @default.
- W4386072088 cites W2101105183 @default.
- W4386072088 cites W2108598243 @default.
- W4386072088 cites W2133459682 @default.
- W4386072088 cites W2194775991 @default.
- W4386072088 cites W2467281799 @default.
- W4386072088 cites W2474574787 @default.
- W4386072088 cites W2493181180 @default.
- W4386072088 cites W2507296351 @default.
- W4386072088 cites W2546190447 @default.
- W4386072088 cites W2588822708 @default.
- W4386072088 cites W2607579284 @default.
- W4386072088 cites W2776402438 @default.
- W4386072088 cites W2888894220 @default.
- W4386072088 cites W2938500406 @default.
- W4386072088 cites W2963040148 @default.
- W4386072088 cites W2963163163 @default.
- W4386072088 cites W2963207848 @default.
- W4386072088 cites W2970013683 @default.
- W4386072088 cites W2982553922 @default.
- W4386072088 cites W2994818707 @default.
- W4386072088 cites W3034603197 @default.
- W4386072088 cites W3034655362 @default.
- W4386072088 cites W3035058753 @default.
- W4386072088 cites W3099716136 @default.
- W4386072088 cites W3101029400 @default.
- W4386072088 cites W3104279398 @default.
- W4386072088 cites W3174930619 @default.
- W4386072088 cites W3176372702 @default.
- W4386072088 cites W3177343422 @default.
- W4386072088 cites W3181089721 @default.
- W4386072088 cites W3181252431 @default.
- W4386072088 cites W3182937942 @default.
- W4386072088 cites W3196936439 @default.
- W4386072088 cites W3204331734 @default.
- W4386072088 cites W4220726865 @default.
- W4386072088 cites W4230981054 @default.
- W4386072088 cites W4312244961 @default.
- W4386072088 cites W4312987526 @default.
- W4386072088 cites W4313067071 @default.
- W4386072088 cites W4321512533 @default.
- W4386072088 cites W4386066418 @default.
- W4386072088 cites W4386075743 @default.
- W4386072088 cites W4386075758 @default.
- W4386072088 cites W4386076001 @default.
- W4386072088 cites W4386076063 @default.
- W4386072088 doi "https://doi.org/10.1109/cvpr52729.2023.01056" @default.
- W4386072088 hasPublicationYear "2023" @default.
- W4386072088 type Work @default.
- W4386072088 citedByCount "3" @default.
- W4386072088 countsByYear W43860720882023 @default.
- W4386072088 crossrefType "proceedings-article" @default.
- W4386072088 hasAuthorship W4386072088A5003628834 @default.
- W4386072088 hasAuthorship W4386072088A5010062829 @default.
- W4386072088 hasAuthorship W4386072088A5014436524 @default.
- W4386072088 hasAuthorship W4386072088A5043791403 @default.
- W4386072088 hasAuthorship W4386072088A5046046128 @default.
- W4386072088 hasAuthorship W4386072088A5085178555 @default.
- W4386072088 hasConcept C11413529 @default.
- W4386072088 hasConcept C144024400 @default.
- W4386072088 hasConcept C154945302 @default.
- W4386072088 hasConcept C202269582 @default.
- W4386072088 hasConcept C204321447 @default.
- W4386072088 hasConcept C23123220 @default.
- W4386072088 hasConcept C2779231336 @default.
- W4386072088 hasConcept C2779903281 @default.
- W4386072088 hasConcept C2780226545 @default.
- W4386072088 hasConcept C36289849 @default.
- W4386072088 hasConcept C41008148 @default.
- W4386072088 hasConcept C41608201 @default.
- W4386072088 hasConcept C54355233 @default.
- W4386072088 hasConcept C86803240 @default.
- W4386072088 hasConcept C97931131 @default.
- W4386072088 hasConceptScore W4386072088C11413529 @default.
- W4386072088 hasConceptScore W4386072088C144024400 @default.
- W4386072088 hasConceptScore W4386072088C154945302 @default.
- W4386072088 hasConceptScore W4386072088C202269582 @default.
- W4386072088 hasConceptScore W4386072088C204321447 @default.
- W4386072088 hasConceptScore W4386072088C23123220 @default.
- W4386072088 hasConceptScore W4386072088C2779231336 @default.
- W4386072088 hasConceptScore W4386072088C2779903281 @default.
- W4386072088 hasConceptScore W4386072088C2780226545 @default.
- W4386072088 hasConceptScore W4386072088C36289849 @default.
- W4386072088 hasConceptScore W4386072088C41008148 @default.
- W4386072088 hasConceptScore W4386072088C41608201 @default.
- W4386072088 hasConceptScore W4386072088C54355233 @default.
- W4386072088 hasConceptScore W4386072088C86803240 @default.
- W4386072088 hasConceptScore W4386072088C97931131 @default.
- W4386072088 hasLocation W43860720881 @default.