Matches in SemOpenAlex for { <https://semopenalex.org/work/W4372259802> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W4372259802 abstract "Heterogeneous graphs provide a compact, efficient, and scalable way to model data involving multiple disparate modalities. This makes modeling audiovisual data using heterogeneous graphs an attractive option. However, graph structure does not appear naturally in audiovisual data. Graphs for audiovisual data are constructed manually which is both difficult and sub-optimal. In this work, we address this problem by (i) proposing a parametric graph construction strategy for the intra-modal edges, and (ii) learning the crossmodal edges. To this end, we develop a new model, heterogeneous graph crossmodal network (HGCN) that learns the crossmodal edges. Our proposed model can adapt to various spatial and temporal scales owing to its parametric construction, while the learnable crossmodal edges effectively connect the relevant nodes across modalities. Experiments on a large benchmark dataset (AudioSet) show that our model is state-of-the-art (0.53 mean average precision), outperforming transformer-based models and other graph-based models. Our code is available at github.com/AmirSh15/Crossmodalitygraph" @default.
- W4372259802 created "2023-05-07" @default.
- W4372259802 creator A5008305849 @default.
- W4372259802 creator A5015290567 @default.
- W4372259802 creator A5021354054 @default.
- W4372259802 creator A5080384065 @default.
- W4372259802 date "2023-06-04" @default.
- W4372259802 modified "2023-10-18" @default.
- W4372259802 title "Heterogeneous Graph Learning for Acoustic Event Classification" @default.
- W4372259802 cites W2015394094 @default.
- W4372259802 cites W2529337537 @default.
- W4372259802 cites W2950081742 @default.
- W4372259802 cites W2963155035 @default.
- W4372259802 cites W3004507689 @default.
- W4372259802 cites W3012871709 @default.
- W4372259802 cites W3094550259 @default.
- W4372259802 cites W3162322471 @default.
- W4372259802 cites W3162387328 @default.
- W4372259802 cites W3162391496 @default.
- W4372259802 cites W3174517569 @default.
- W4372259802 cites W3176349206 @default.
- W4372259802 cites W3196974791 @default.
- W4372259802 cites W4285414370 @default.
- W4372259802 cites W4297841792 @default.
- W4372259802 cites W4298357738 @default.
- W4372259802 doi "https://doi.org/10.1109/icassp49357.2023.10095073" @default.
- W4372259802 hasPublicationYear "2023" @default.
- W4372259802 type Work @default.
- W4372259802 citedByCount "0" @default.
- W4372259802 crossrefType "proceedings-article" @default.
- W4372259802 hasAuthorship W4372259802A5008305849 @default.
- W4372259802 hasAuthorship W4372259802A5015290567 @default.
- W4372259802 hasAuthorship W4372259802A5021354054 @default.
- W4372259802 hasAuthorship W4372259802A5080384065 @default.
- W4372259802 hasBestOaLocation W43722598021 @default.
- W4372259802 hasConcept C105795698 @default.
- W4372259802 hasConcept C117251300 @default.
- W4372259802 hasConcept C119857082 @default.
- W4372259802 hasConcept C132525143 @default.
- W4372259802 hasConcept C144024400 @default.
- W4372259802 hasConcept C154945302 @default.
- W4372259802 hasConcept C169760540 @default.
- W4372259802 hasConcept C178253425 @default.
- W4372259802 hasConcept C26760741 @default.
- W4372259802 hasConcept C2779903281 @default.
- W4372259802 hasConcept C33923547 @default.
- W4372259802 hasConcept C36289849 @default.
- W4372259802 hasConcept C41008148 @default.
- W4372259802 hasConcept C48044578 @default.
- W4372259802 hasConcept C60115397 @default.
- W4372259802 hasConcept C67186912 @default.
- W4372259802 hasConcept C77088390 @default.
- W4372259802 hasConcept C80444323 @default.
- W4372259802 hasConcept C86803240 @default.
- W4372259802 hasConceptScore W4372259802C105795698 @default.
- W4372259802 hasConceptScore W4372259802C117251300 @default.
- W4372259802 hasConceptScore W4372259802C119857082 @default.
- W4372259802 hasConceptScore W4372259802C132525143 @default.
- W4372259802 hasConceptScore W4372259802C144024400 @default.
- W4372259802 hasConceptScore W4372259802C154945302 @default.
- W4372259802 hasConceptScore W4372259802C169760540 @default.
- W4372259802 hasConceptScore W4372259802C178253425 @default.
- W4372259802 hasConceptScore W4372259802C26760741 @default.
- W4372259802 hasConceptScore W4372259802C2779903281 @default.
- W4372259802 hasConceptScore W4372259802C33923547 @default.
- W4372259802 hasConceptScore W4372259802C36289849 @default.
- W4372259802 hasConceptScore W4372259802C41008148 @default.
- W4372259802 hasConceptScore W4372259802C48044578 @default.
- W4372259802 hasConceptScore W4372259802C60115397 @default.
- W4372259802 hasConceptScore W4372259802C67186912 @default.
- W4372259802 hasConceptScore W4372259802C77088390 @default.
- W4372259802 hasConceptScore W4372259802C80444323 @default.
- W4372259802 hasConceptScore W4372259802C86803240 @default.
- W4372259802 hasLocation W43722598021 @default.
- W4372259802 hasLocation W43722598022 @default.
- W4372259802 hasLocation W43722598023 @default.
- W4372259802 hasOpenAccess W4372259802 @default.
- W4372259802 hasPrimaryLocation W43722598021 @default.
- W4372259802 hasRelatedWork W1525643724 @default.
- W4372259802 hasRelatedWork W2067938758 @default.
- W4372259802 hasRelatedWork W2302028273 @default.
- W4372259802 hasRelatedWork W2333420780 @default.
- W4372259802 hasRelatedWork W2345652077 @default.
- W4372259802 hasRelatedWork W2364921833 @default.
- W4372259802 hasRelatedWork W2382623646 @default.
- W4372259802 hasRelatedWork W2961085424 @default.
- W4372259802 hasRelatedWork W3087771547 @default.
- W4372259802 hasRelatedWork W4313547873 @default.
- W4372259802 isParatext "false" @default.
- W4372259802 isRetracted "false" @default.
- W4372259802 workType "article" @default.