Matches in SemOpenAlex for { <https://semopenalex.org/work/W2981660636> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W2981660636 abstract "Acting as a bridge between vision and language, visual relationship detection (VRD) aims to represent objects and their interactions in an image with several relationship triplets. Nevertheless, the conventional VRD task shows little consideration for the penalization of incorrect relationship predictions, which in turn undermines its support for image understanding applications. In this paper, we propose a novel VRD task named hierarchical visual relationship detection (HVRD), which encourages predictions with abstract yet compatible relationship triplets when the confidence level of the specific image content is relatively low. Meanwhile, HVRD can handle the inevitable ambiguity of groundtruth annotation in VRD. Based on this, we propose a HVRD method, consisting of hierarchical object detection and hierarchical predicate detection. It can effectively detect the hierarchical visual relationships by exploiting both object concept hierarchy and predicate concept hierarchy with order embedding. We also propose the first datasets for HVRD evaluation, H-VRD and H-VG, by expanding the relationship category spaces of VRD and VG datasets to hierarchical ones respectively. The experimental results show that our method is superior to the state-of-the-art baselines." @default.
- W2981660636 created "2019-11-01" @default.
- W2981660636 creator A5006497264 @default.
- W2981660636 creator A5021809690 @default.
- W2981660636 creator A5035112538 @default.
- W2981660636 creator A5070294325 @default.
- W2981660636 creator A5084591510 @default.
- W2981660636 date "2019-10-15" @default.
- W2981660636 modified "2023-10-18" @default.
- W2981660636 title "Hierarchical Visual Relationship Detection" @default.
- W2981660636 cites W1498158520 @default.
- W2981660636 cites W1536680647 @default.
- W2981660636 cites W1933349210 @default.
- W2981660636 cites W1959000896 @default.
- W2981660636 cites W2049705550 @default.
- W2981660636 cites W2081580037 @default.
- W2981660636 cites W2089060621 @default.
- W2981660636 cites W2105101328 @default.
- W2981660636 cites W2108598243 @default.
- W2981660636 cites W2135166986 @default.
- W2981660636 cites W2143017621 @default.
- W2981660636 cites W2277195237 @default.
- W2981660636 cites W2340690086 @default.
- W2981660636 cites W2479423890 @default.
- W2981660636 cites W2556388456 @default.
- W2981660636 cites W2570343428 @default.
- W2981660636 cites W2579549467 @default.
- W2981660636 cites W2591644541 @default.
- W2981660636 cites W2604254608 @default.
- W2981660636 cites W2607855566 @default.
- W2981660636 cites W2765137706 @default.
- W2981660636 cites W2777602943 @default.
- W2981660636 cites W2799029629 @default.
- W2981660636 cites W2883170015 @default.
- W2981660636 cites W2886970679 @default.
- W2981660636 cites W2887029921 @default.
- W2981660636 cites W2889811054 @default.
- W2981660636 cites W2890531016 @default.
- W2981660636 cites W2936758701 @default.
- W2981660636 cites W2962737704 @default.
- W2981660636 cites W2963536419 @default.
- W2981660636 cites W2963649796 @default.
- W2981660636 cites W2963650529 @default.
- W2981660636 cites W2963980128 @default.
- W2981660636 cites W3005285208 @default.
- W2981660636 cites W64813323 @default.
- W2981660636 doi "https://doi.org/10.1145/3343031.3350921" @default.
- W2981660636 hasPublicationYear "2019" @default.
- W2981660636 type Work @default.
- W2981660636 sameAs 2981660636 @default.
- W2981660636 citedByCount "5" @default.
- W2981660636 countsByYear W29816606362020 @default.
- W2981660636 countsByYear W29816606362021 @default.
- W2981660636 countsByYear W29816606362022 @default.
- W2981660636 crossrefType "proceedings-article" @default.
- W2981660636 hasAuthorship W2981660636A5006497264 @default.
- W2981660636 hasAuthorship W2981660636A5021809690 @default.
- W2981660636 hasAuthorship W2981660636A5035112538 @default.
- W2981660636 hasAuthorship W2981660636A5070294325 @default.
- W2981660636 hasAuthorship W2981660636A5084591510 @default.
- W2981660636 hasConcept C154945302 @default.
- W2981660636 hasConcept C31972630 @default.
- W2981660636 hasConcept C41008148 @default.
- W2981660636 hasConceptScore W2981660636C154945302 @default.
- W2981660636 hasConceptScore W2981660636C31972630 @default.
- W2981660636 hasConceptScore W2981660636C41008148 @default.
- W2981660636 hasLocation W29816606361 @default.
- W2981660636 hasOpenAccess W2981660636 @default.
- W2981660636 hasPrimaryLocation W29816606361 @default.
- W2981660636 hasRelatedWork W1891287906 @default.
- W2981660636 hasRelatedWork W1969923398 @default.
- W2981660636 hasRelatedWork W2036807459 @default.
- W2981660636 hasRelatedWork W2058170566 @default.
- W2981660636 hasRelatedWork W2170022336 @default.
- W2981660636 hasRelatedWork W2229312674 @default.
- W2981660636 hasRelatedWork W258625772 @default.
- W2981660636 hasRelatedWork W2755342338 @default.
- W2981660636 hasRelatedWork W2772917594 @default.
- W2981660636 hasRelatedWork W3116076068 @default.
- W2981660636 isParatext "false" @default.
- W2981660636 isRetracted "false" @default.
- W2981660636 magId "2981660636" @default.
- W2981660636 workType "article" @default.