Matches in SemOpenAlex for { <https://semopenalex.org/work/W3206675006> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W3206675006 abstract "Video question answering~(Video-QA) is a task of answering a natural language question related to the content of a video. Existing methods generally explore the single interactions between objects or between frames, which are insufficient to deal with the sophisticated scenes in videos. To tackle this problem, we propose a novel model, termed Progressive Graph Attention Network (PGAT), which can jointly explore the multiple visual relations on object-level, frame-level and clip-level. Specifically, in the object-level relation encoding, we design two kinds of complementary graphs, one for learning the spatial and semantic relations between objects from the same frame, the other for modeling the temporal relations between the same object from different frames. The frame-level graph explores the interactions between diverse frames to record the fine-grained appearance change, while the clip-level graph models the temporal and semantic relations between various actions from clips. These different-level graphs are concatenated in a progressive manner to learn the visual relations from low-level to high-level. Furthermore, we for the first time identified that there are serious answer biases with TGIF-QA, a very large Video-QA dataset, and reconstructed a new dataset based on it to overcome the biases, called TGIF-QA-R. We evaluate the proposed model on three benchmark datasets and the new TGIF-QA-R, and the experimental results demonstrate that our model significantly outperforms other state-of-the-art models. Our codes and dataset are available at https://github.com/PengLiang-cn/PGAT." @default.
- W3206675006 created "2021-10-25" @default.
- W3206675006 creator A5011057377 @default.
- W3206675006 creator A5024408423 @default.
- W3206675006 creator A5063303076 @default.
- W3206675006 creator A5083003270 @default.
- W3206675006 date "2021-10-17" @default.
- W3206675006 modified "2023-10-11" @default.
- W3206675006 title "Progressive Graph Attention Network for Video Question Answering" @default.
- W3206675006 cites W1933349210 @default.
- W3206675006 cites W2064675550 @default.
- W3206675006 cites W2142192571 @default.
- W3206675006 cites W2560730294 @default.
- W3206675006 cites W2755385356 @default.
- W3206675006 cites W2766237723 @default.
- W3206675006 cites W2904323922 @default.
- W3206675006 cites W2904452845 @default.
- W3206675006 cites W2948358897 @default.
- W3206675006 cites W2962934715 @default.
- W3206675006 cites W2963954913 @default.
- W3206675006 cites W2964345214 @default.
- W3206675006 cites W2981578638 @default.
- W3206675006 cites W2981582341 @default.
- W3206675006 cites W2981985547 @default.
- W3206675006 cites W2997344006 @default.
- W3206675006 cites W2997805943 @default.
- W3206675006 cites W2998166190 @default.
- W3206675006 cites W3135856559 @default.
- W3206675006 cites W3169472988 @default.
- W3206675006 doi "https://doi.org/10.1145/3474085.3475193" @default.
- W3206675006 hasPublicationYear "2021" @default.
- W3206675006 type Work @default.
- W3206675006 sameAs 3206675006 @default.
- W3206675006 citedByCount "10" @default.
- W3206675006 countsByYear W32066750062022 @default.
- W3206675006 countsByYear W32066750062023 @default.
- W3206675006 crossrefType "proceedings-article" @default.
- W3206675006 hasAuthorship W3206675006A5011057377 @default.
- W3206675006 hasAuthorship W3206675006A5024408423 @default.
- W3206675006 hasAuthorship W3206675006A5063303076 @default.
- W3206675006 hasAuthorship W3206675006A5083003270 @default.
- W3206675006 hasConcept C119857082 @default.
- W3206675006 hasConcept C126042441 @default.
- W3206675006 hasConcept C132525143 @default.
- W3206675006 hasConcept C13280743 @default.
- W3206675006 hasConcept C154945302 @default.
- W3206675006 hasConcept C179372163 @default.
- W3206675006 hasConcept C185798385 @default.
- W3206675006 hasConcept C204321447 @default.
- W3206675006 hasConcept C205649164 @default.
- W3206675006 hasConcept C205711294 @default.
- W3206675006 hasConcept C23123220 @default.
- W3206675006 hasConcept C27511587 @default.
- W3206675006 hasConcept C2781238097 @default.
- W3206675006 hasConcept C41008148 @default.
- W3206675006 hasConcept C44291984 @default.
- W3206675006 hasConcept C76155785 @default.
- W3206675006 hasConcept C80444323 @default.
- W3206675006 hasConceptScore W3206675006C119857082 @default.
- W3206675006 hasConceptScore W3206675006C126042441 @default.
- W3206675006 hasConceptScore W3206675006C132525143 @default.
- W3206675006 hasConceptScore W3206675006C13280743 @default.
- W3206675006 hasConceptScore W3206675006C154945302 @default.
- W3206675006 hasConceptScore W3206675006C179372163 @default.
- W3206675006 hasConceptScore W3206675006C185798385 @default.
- W3206675006 hasConceptScore W3206675006C204321447 @default.
- W3206675006 hasConceptScore W3206675006C205649164 @default.
- W3206675006 hasConceptScore W3206675006C205711294 @default.
- W3206675006 hasConceptScore W3206675006C23123220 @default.
- W3206675006 hasConceptScore W3206675006C27511587 @default.
- W3206675006 hasConceptScore W3206675006C2781238097 @default.
- W3206675006 hasConceptScore W3206675006C41008148 @default.
- W3206675006 hasConceptScore W3206675006C44291984 @default.
- W3206675006 hasConceptScore W3206675006C76155785 @default.
- W3206675006 hasConceptScore W3206675006C80444323 @default.
- W3206675006 hasFunder F4320321001 @default.
- W3206675006 hasFunder F4320321543 @default.
- W3206675006 hasFunder F4320335787 @default.
- W3206675006 hasLocation W32066750061 @default.
- W3206675006 hasOpenAccess W3206675006 @default.
- W3206675006 hasPrimaryLocation W32066750061 @default.
- W3206675006 hasRelatedWork W128392744 @default.
- W3206675006 hasRelatedWork W1978557720 @default.
- W3206675006 hasRelatedWork W207304934 @default.
- W3206675006 hasRelatedWork W2135033253 @default.
- W3206675006 hasRelatedWork W2233955765 @default.
- W3206675006 hasRelatedWork W2747680751 @default.
- W3206675006 hasRelatedWork W2798526799 @default.
- W3206675006 hasRelatedWork W3107474891 @default.
- W3206675006 hasRelatedWork W4287898599 @default.
- W3206675006 hasRelatedWork W2153799799 @default.
- W3206675006 isParatext "false" @default.
- W3206675006 isRetracted "false" @default.
- W3206675006 magId "3206675006" @default.
- W3206675006 workType "article" @default.