Matches in SemOpenAlex for { <https://semopenalex.org/work/W3207922251> ?p ?o ?g. }
Showing items 1 to 72 of
72
with 100 items per page.
- W3207922251 abstract "In this talk we present recent progress on large-scale learning of multimodal video representations. We start by presenting VideoBert, a joint model for video and language, repurposing the Bert model for multimodal data. This model achieves state-of-the-art results on zero shot prediction and video captioning. Next we show how to extend learning from instruction videos to general movies based on cross-modal supervision. We use movie screenplays to learn a speech to action classifiers and use these classifiers to mine video clips from thousands of hours of movies. We demonstrate a performance comparable or better than fully supervised approaches for action classification. Next we present an approach for video question answering which relies on training from instruction videos and cross-modal supervision with a textual question answer module. We show state-of-the-art results for video question answering without any supervision (zero-shot VQA) and demonstrate that our approach obtains competitive results for pre-training and then fine-tuning on video question answering datasets. We conclude our talk by presenting a recent video feature which is fully transformer based. Our Video Vision Transformer (ViViT) is shown to outperform the state-of-the-art on video classification. Furthermore, it is flexible and allows for performance / accuracy trade-off based on several different architectures." @default.
- W3207922251 created "2021-10-25" @default.
- W3207922251 creator A5045217258 @default.
- W3207922251 date "2021-10-17" @default.
- W3207922251 modified "2023-09-27" @default.
- W3207922251 title "Do you see what I see?" @default.
- W3207922251 doi "https://doi.org/10.1145/3474085.3476967" @default.
- W3207922251 hasPublicationYear "2021" @default.
- W3207922251 type Work @default.
- W3207922251 sameAs 3207922251 @default.
- W3207922251 citedByCount "0" @default.
- W3207922251 crossrefType "proceedings-article" @default.
- W3207922251 hasAuthorship W3207922251A5045217258 @default.
- W3207922251 hasConcept C115961682 @default.
- W3207922251 hasConcept C119857082 @default.
- W3207922251 hasConcept C121332964 @default.
- W3207922251 hasConcept C154945302 @default.
- W3207922251 hasConcept C157657479 @default.
- W3207922251 hasConcept C165801399 @default.
- W3207922251 hasConcept C185592680 @default.
- W3207922251 hasConcept C188027245 @default.
- W3207922251 hasConcept C18903297 @default.
- W3207922251 hasConcept C2778739407 @default.
- W3207922251 hasConcept C28490314 @default.
- W3207922251 hasConcept C41008148 @default.
- W3207922251 hasConcept C44291984 @default.
- W3207922251 hasConcept C49774154 @default.
- W3207922251 hasConcept C519536355 @default.
- W3207922251 hasConcept C59404180 @default.
- W3207922251 hasConcept C62520636 @default.
- W3207922251 hasConcept C66322947 @default.
- W3207922251 hasConcept C71139939 @default.
- W3207922251 hasConcept C774472 @default.
- W3207922251 hasConcept C86803240 @default.
- W3207922251 hasConceptScore W3207922251C115961682 @default.
- W3207922251 hasConceptScore W3207922251C119857082 @default.
- W3207922251 hasConceptScore W3207922251C121332964 @default.
- W3207922251 hasConceptScore W3207922251C154945302 @default.
- W3207922251 hasConceptScore W3207922251C157657479 @default.
- W3207922251 hasConceptScore W3207922251C165801399 @default.
- W3207922251 hasConceptScore W3207922251C185592680 @default.
- W3207922251 hasConceptScore W3207922251C188027245 @default.
- W3207922251 hasConceptScore W3207922251C18903297 @default.
- W3207922251 hasConceptScore W3207922251C2778739407 @default.
- W3207922251 hasConceptScore W3207922251C28490314 @default.
- W3207922251 hasConceptScore W3207922251C41008148 @default.
- W3207922251 hasConceptScore W3207922251C44291984 @default.
- W3207922251 hasConceptScore W3207922251C49774154 @default.
- W3207922251 hasConceptScore W3207922251C519536355 @default.
- W3207922251 hasConceptScore W3207922251C59404180 @default.
- W3207922251 hasConceptScore W3207922251C62520636 @default.
- W3207922251 hasConceptScore W3207922251C66322947 @default.
- W3207922251 hasConceptScore W3207922251C71139939 @default.
- W3207922251 hasConceptScore W3207922251C774472 @default.
- W3207922251 hasConceptScore W3207922251C86803240 @default.
- W3207922251 hasLocation W32079222511 @default.
- W3207922251 hasOpenAccess W3207922251 @default.
- W3207922251 hasPrimaryLocation W32079222511 @default.
- W3207922251 hasRelatedWork W1499754964 @default.
- W3207922251 hasRelatedWork W2975706270 @default.
- W3207922251 hasRelatedWork W2997591391 @default.
- W3207922251 hasRelatedWork W3025796084 @default.
- W3207922251 hasRelatedWork W3128093669 @default.
- W3207922251 hasRelatedWork W3138074355 @default.
- W3207922251 hasRelatedWork W3198730297 @default.
- W3207922251 hasRelatedWork W3207922251 @default.
- W3207922251 hasRelatedWork W4287777632 @default.
- W3207922251 hasRelatedWork W4320561490 @default.
- W3207922251 isParatext "false" @default.
- W3207922251 isRetracted "false" @default.
- W3207922251 magId "3207922251" @default.
- W3207922251 workType "article" @default.