Matches in SemOpenAlex for { <https://semopenalex.org/work/W4319300107> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W4319300107 abstract "Scene graph generation (SGG) models seek to detect relationships between objects in a given image. One challenge in this area is the biased distribution of predicates in the dataset and the semantic space. Recent works incorporating knowledge graphs with scene graphs prove effective in improving recall for the tail predicate classes. Moreover, many recent SGG approaches with promising results explicitly redistribute the predicates in both the training process and in the prediction step. To incorporate external knowledge, we construct a commonsense knowledge graph by integrating ConceptNet and Wikidata. To explicitly unbias SGG with knowledge in the reasoning process, we propose a novel framework, Explicit Ontological Adjustment (EOA), to adjust the graph model predictions with knowledge priors. We use the edge matrix from the commonsense knowledge graph as a module in the graph neural network model to refine the relationship detection process. This module proves effective in alleviating the long-tail distribution of predicates. When combined, we show that these modules achieve state-of-the-art performance on the Visual Genome dataset in most cases. The source code is available at https://github.com/zhanwenchen/eoa." @default.
- W4319300107 created "2023-02-07" @default.
- W4319300107 creator A5016719957 @default.
- W4319300107 creator A5017784461 @default.
- W4319300107 creator A5026512168 @default.
- W4319300107 date "2023-01-01" @default.
- W4319300107 modified "2023-10-02" @default.
- W4319300107 title "More Knowledge, Less Bias: Unbiasing Scene Graph Generation with Explicit Ontological Adjustment" @default.
- W4319300107 cites W2077069816 @default.
- W4319300107 cites W2080133951 @default.
- W4319300107 cites W2090243146 @default.
- W4319300107 cites W2107658650 @default.
- W4319300107 cites W2194775991 @default.
- W4319300107 cites W2250378130 @default.
- W4319300107 cites W2277195237 @default.
- W4319300107 cites W2560920409 @default.
- W4319300107 cites W2565639579 @default.
- W4319300107 cites W2579549467 @default.
- W4319300107 cites W2740432174 @default.
- W4319300107 cites W2800046969 @default.
- W4319300107 cites W2950096400 @default.
- W4319300107 cites W2950339735 @default.
- W4319300107 cites W2962785943 @default.
- W4319300107 cites W2963048642 @default.
- W4319300107 cites W2963101081 @default.
- W4319300107 cites W2963101956 @default.
- W4319300107 cites W2963184176 @default.
- W4319300107 cites W2963486920 @default.
- W4319300107 cites W2963536419 @default.
- W4319300107 cites W2963938081 @default.
- W4319300107 cites W2964199361 @default.
- W4319300107 cites W2964303913 @default.
- W4319300107 cites W2969679616 @default.
- W4319300107 cites W3035017890 @default.
- W4319300107 cites W3104748221 @default.
- W4319300107 cites W3108864070 @default.
- W4319300107 cites W3173399746 @default.
- W4319300107 cites W3191092153 @default.
- W4319300107 cites W3196936439 @default.
- W4319300107 cites W3205051885 @default.
- W4319300107 cites W4214693531 @default.
- W4319300107 cites W4292025450 @default.
- W4319300107 cites W639708223 @default.
- W4319300107 doi "https://doi.org/10.1109/wacv56688.2023.00401" @default.
- W4319300107 hasPublicationYear "2023" @default.
- W4319300107 type Work @default.
- W4319300107 citedByCount "1" @default.
- W4319300107 countsByYear W43193001072023 @default.
- W4319300107 crossrefType "proceedings-article" @default.
- W4319300107 hasAuthorship W4319300107A5016719957 @default.
- W4319300107 hasAuthorship W4319300107A5017784461 @default.
- W4319300107 hasAuthorship W4319300107A5026512168 @default.
- W4319300107 hasConcept C107673813 @default.
- W4319300107 hasConcept C120567893 @default.
- W4319300107 hasConcept C132525143 @default.
- W4319300107 hasConcept C140146324 @default.
- W4319300107 hasConcept C154945302 @default.
- W4319300107 hasConcept C177769412 @default.
- W4319300107 hasConcept C179372163 @default.
- W4319300107 hasConcept C199360897 @default.
- W4319300107 hasConcept C205711294 @default.
- W4319300107 hasConcept C2987255567 @default.
- W4319300107 hasConcept C30542707 @default.
- W4319300107 hasConcept C41008148 @default.
- W4319300107 hasConcept C80444323 @default.
- W4319300107 hasConcept C81669768 @default.
- W4319300107 hasConceptScore W4319300107C107673813 @default.
- W4319300107 hasConceptScore W4319300107C120567893 @default.
- W4319300107 hasConceptScore W4319300107C132525143 @default.
- W4319300107 hasConceptScore W4319300107C140146324 @default.
- W4319300107 hasConceptScore W4319300107C154945302 @default.
- W4319300107 hasConceptScore W4319300107C177769412 @default.
- W4319300107 hasConceptScore W4319300107C179372163 @default.
- W4319300107 hasConceptScore W4319300107C199360897 @default.
- W4319300107 hasConceptScore W4319300107C205711294 @default.
- W4319300107 hasConceptScore W4319300107C2987255567 @default.
- W4319300107 hasConceptScore W4319300107C30542707 @default.
- W4319300107 hasConceptScore W4319300107C41008148 @default.
- W4319300107 hasConceptScore W4319300107C80444323 @default.
- W4319300107 hasConceptScore W4319300107C81669768 @default.
- W4319300107 hasFunder F4320316514 @default.
- W4319300107 hasLocation W43193001071 @default.
- W4319300107 hasOpenAccess W4319300107 @default.
- W4319300107 hasPrimaryLocation W43193001071 @default.
- W4319300107 hasRelatedWork W2923818335 @default.
- W4319300107 hasRelatedWork W2953365189 @default.
- W4319300107 hasRelatedWork W3006657421 @default.
- W4319300107 hasRelatedWork W3094328377 @default.
- W4319300107 hasRelatedWork W3097486074 @default.
- W4319300107 hasRelatedWork W3113685523 @default.
- W4319300107 hasRelatedWork W4206461159 @default.
- W4319300107 hasRelatedWork W4285208802 @default.
- W4319300107 hasRelatedWork W4302773889 @default.
- W4319300107 hasRelatedWork W4385848775 @default.
- W4319300107 isParatext "false" @default.
- W4319300107 isRetracted "false" @default.
- W4319300107 workType "article" @default.