Matches in SemOpenAlex for { <https://semopenalex.org/work/W2759653627> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W2759653627 abstract "Attention mechanisms have been widely applied in the Visual Question Answering (VQA) task, as they help to focus on the area-of-interest of both visual and textual information. To answer the questions correctly, the model needs to selectively target different areas of an image, which suggests that an attention-based model may benefit from an explicit attention supervision. In this work, we aim to address the problem of adding attention supervision to VQA models. Since there is a lack of human attention data, we first propose a Human Attention Network (HAN) to generate human-like attention maps, training on a recently released dataset called Human ATtention Dataset (VQA-HAT). Then, we apply the pre-trained HAN on the VQA v2.0 dataset to automatically produce the human-like attention maps for all image-question pairs. The generated human-like attention map dataset for the VQA v2.0 dataset is named as Human-Like ATtention (HLAT) dataset. Finally, we apply human-like attention supervision to an attention-based VQA model. The experiments show that adding human-like supervision yields a more accurate attention together with a better performance, showing a promising future for human-like attention supervision in VQA." @default.
- W2759653627 created "2017-10-06" @default.
- W2759653627 creator A5008417257 @default.
- W2759653627 creator A5011304041 @default.
- W2759653627 creator A5087916440 @default.
- W2759653627 date "2018-04-27" @default.
- W2759653627 modified "2023-10-18" @default.
- W2759653627 title "Exploring Human-Like Attention Supervision in Visual Question Answering" @default.
- W2759653627 cites W1522301498 @default.
- W2759653627 cites W1934890906 @default.
- W2759653627 cites W2108598243 @default.
- W2759653627 cites W2157331557 @default.
- W2759653627 cites W2190656909 @default.
- W2759653627 cites W2194775991 @default.
- W2759653627 cites W2410323755 @default.
- W2759653627 cites W2412393473 @default.
- W2759653627 cites W2412400526 @default.
- W2759653627 cites W2424413254 @default.
- W2759653627 cites W2463565445 @default.
- W2759653627 cites W2737056415 @default.
- W2759653627 cites W2747148117 @default.
- W2759653627 cites W2949197413 @default.
- W2759653627 cites W2950726992 @default.
- W2759653627 cites W2950761309 @default.
- W2759653627 cites W2951690276 @default.
- W2759653627 cites W2952246170 @default.
- W2759653627 cites W2963191264 @default.
- W2759653627 cites W2963954913 @default.
- W2759653627 cites W2964091467 @default.
- W2759653627 cites W2964118342 @default.
- W2759653627 cites W2964138017 @default.
- W2759653627 cites W753012316 @default.
- W2759653627 doi "https://doi.org/10.1609/aaai.v32i1.12272" @default.
- W2759653627 hasPublicationYear "2018" @default.
- W2759653627 type Work @default.
- W2759653627 sameAs 2759653627 @default.
- W2759653627 citedByCount "45" @default.
- W2759653627 countsByYear W27596536272018 @default.
- W2759653627 countsByYear W27596536272019 @default.
- W2759653627 countsByYear W27596536272020 @default.
- W2759653627 countsByYear W27596536272021 @default.
- W2759653627 countsByYear W27596536272022 @default.
- W2759653627 countsByYear W27596536272023 @default.
- W2759653627 crossrefType "journal-article" @default.
- W2759653627 hasAuthorship W2759653627A5008417257 @default.
- W2759653627 hasAuthorship W2759653627A5011304041 @default.
- W2759653627 hasAuthorship W2759653627A5087916440 @default.
- W2759653627 hasBestOaLocation W27596536271 @default.
- W2759653627 hasConcept C119857082 @default.
- W2759653627 hasConcept C120665830 @default.
- W2759653627 hasConcept C121332964 @default.
- W2759653627 hasConcept C154945302 @default.
- W2759653627 hasConcept C15744967 @default.
- W2759653627 hasConcept C162324750 @default.
- W2759653627 hasConcept C169760540 @default.
- W2759653627 hasConcept C169900460 @default.
- W2759653627 hasConcept C187736073 @default.
- W2759653627 hasConcept C192209626 @default.
- W2759653627 hasConcept C2522767166 @default.
- W2759653627 hasConcept C2780451532 @default.
- W2759653627 hasConcept C2986089797 @default.
- W2759653627 hasConcept C2993807640 @default.
- W2759653627 hasConcept C41008148 @default.
- W2759653627 hasConcept C44291984 @default.
- W2759653627 hasConceptScore W2759653627C119857082 @default.
- W2759653627 hasConceptScore W2759653627C120665830 @default.
- W2759653627 hasConceptScore W2759653627C121332964 @default.
- W2759653627 hasConceptScore W2759653627C154945302 @default.
- W2759653627 hasConceptScore W2759653627C15744967 @default.
- W2759653627 hasConceptScore W2759653627C162324750 @default.
- W2759653627 hasConceptScore W2759653627C169760540 @default.
- W2759653627 hasConceptScore W2759653627C169900460 @default.
- W2759653627 hasConceptScore W2759653627C187736073 @default.
- W2759653627 hasConceptScore W2759653627C192209626 @default.
- W2759653627 hasConceptScore W2759653627C2522767166 @default.
- W2759653627 hasConceptScore W2759653627C2780451532 @default.
- W2759653627 hasConceptScore W2759653627C2986089797 @default.
- W2759653627 hasConceptScore W2759653627C2993807640 @default.
- W2759653627 hasConceptScore W2759653627C41008148 @default.
- W2759653627 hasConceptScore W2759653627C44291984 @default.
- W2759653627 hasIssue "1" @default.
- W2759653627 hasLocation W27596536271 @default.
- W2759653627 hasLocation W27596536272 @default.
- W2759653627 hasOpenAccess W2759653627 @default.
- W2759653627 hasPrimaryLocation W27596536271 @default.
- W2759653627 hasRelatedWork W1513194152 @default.
- W2759653627 hasRelatedWork W2734515856 @default.
- W2759653627 hasRelatedWork W2759653627 @default.
- W2759653627 hasRelatedWork W2795782778 @default.
- W2759653627 hasRelatedWork W2809632469 @default.
- W2759653627 hasRelatedWork W2963466731 @default.
- W2759653627 hasRelatedWork W2964305556 @default.
- W2759653627 hasRelatedWork W3038897048 @default.
- W2759653627 hasRelatedWork W3081809434 @default.
- W2759653627 hasRelatedWork W4241812583 @default.
- W2759653627 hasVolume "32" @default.
- W2759653627 isParatext "false" @default.
- W2759653627 isRetracted "false" @default.
- W2759653627 magId "2759653627" @default.
- W2759653627 workType "article" @default.