Matches in SemOpenAlex for { <https://semopenalex.org/work/W2796869301> ?p ?o ?g. }
- W2796869301 abstract "We address the problem of text-based activity retrieval in video. Given a sentence describing an activity, our task is to retrieve matching clips from an untrimmed video. To capture the inherent structures present in both text and video, we introduce a multilevel model that integrates vision and language features earlier and more tightly than prior work. First, we inject text features early on when generating clip proposals, to help eliminate unlikely clips and thus speed up processing and boost performance. Second, to learn a fine-grained similarity metric for retrieval, we use visual features to modulate the processing of query sentences at the word level in a recurrent neural network. A multi-task loss is also employed by adding query re-generation as an auxiliary task. Our approach significantly outperforms prior work on two challenging benchmarks: Charades-STA and ActivityNet Captions." @default.
- W2796869301 created "2018-04-24" @default.
- W2796869301 creator A5001126038 @default.
- W2796869301 creator A5042663603 @default.
- W2796869301 creator A5053011888 @default.
- W2796869301 creator A5075906727 @default.
- W2796869301 creator A5079036364 @default.
- W2796869301 date "2018-04-13" @default.
- W2796869301 modified "2023-09-24" @default.
- W2796869301 title "Text-to-Clip Video Retrieval with Early Fusion and Re-Captioning." @default.
- W2796869301 cites W1514535095 @default.
- W2796869301 cites W1522734439 @default.
- W2796869301 cites W1586939924 @default.
- W2796869301 cites W1797109199 @default.
- W2796869301 cites W1895577753 @default.
- W2796869301 cites W1916445035 @default.
- W2796869301 cites W1923332106 @default.
- W2796869301 cites W1931639407 @default.
- W2796869301 cites W1945129080 @default.
- W2796869301 cites W1947481528 @default.
- W2796869301 cites W2102705755 @default.
- W2796869301 cites W2139501017 @default.
- W2796869301 cites W2153579005 @default.
- W2796869301 cites W2250384498 @default.
- W2796869301 cites W2293453011 @default.
- W2796869301 cites W2337252826 @default.
- W2796869301 cites W2346425926 @default.
- W2796869301 cites W2472970127 @default.
- W2796869301 cites W2519328139 @default.
- W2796869301 cites W2563296158 @default.
- W2796869301 cites W2565656701 @default.
- W2796869301 cites W2593722617 @default.
- W2796869301 cites W2606473278 @default.
- W2796869301 cites W2607119937 @default.
- W2796869301 cites W2611596598 @default.
- W2796869301 cites W2612076919 @default.
- W2796869301 cites W2613718673 @default.
- W2796869301 cites W2755876276 @default.
- W2796869301 cites W2779250840 @default.
- W2796869301 cites W2962790054 @default.
- W2796869301 cites W2962803561 @default.
- W2796869301 cites W2963017553 @default.
- W2796869301 cites W2963245493 @default.
- W2796869301 cites W2963247196 @default.
- W2796869301 cites W2963321993 @default.
- W2796869301 cites W2963371637 @default.
- W2796869301 cites W2963656855 @default.
- W2796869301 cites W2963735856 @default.
- W2796869301 cites W2963758027 @default.
- W2796869301 cites W2963775347 @default.
- W2796869301 cites W2963899908 @default.
- W2796869301 cites W2963954913 @default.
- W2796869301 cites W2964089981 @default.
- W2796869301 cites W2964214371 @default.
- W2796869301 cites W2964216549 @default.
- W2796869301 cites W2964241990 @default.
- W2796869301 cites W3098232790 @default.
- W2796869301 cites W3100481960 @default.
- W2796869301 cites W3106250896 @default.
- W2796869301 hasPublicationYear "2018" @default.
- W2796869301 type Work @default.
- W2796869301 sameAs 2796869301 @default.
- W2796869301 citedByCount "7" @default.
- W2796869301 countsByYear W27968693012019 @default.
- W2796869301 countsByYear W27968693012020 @default.
- W2796869301 crossrefType "posted-content" @default.
- W2796869301 hasAuthorship W2796869301A5001126038 @default.
- W2796869301 hasAuthorship W2796869301A5042663603 @default.
- W2796869301 hasAuthorship W2796869301A5053011888 @default.
- W2796869301 hasAuthorship W2796869301A5075906727 @default.
- W2796869301 hasAuthorship W2796869301A5079036364 @default.
- W2796869301 hasConcept C103278499 @default.
- W2796869301 hasConcept C105795698 @default.
- W2796869301 hasConcept C115961682 @default.
- W2796869301 hasConcept C138885662 @default.
- W2796869301 hasConcept C154945302 @default.
- W2796869301 hasConcept C157657479 @default.
- W2796869301 hasConcept C162324750 @default.
- W2796869301 hasConcept C165064840 @default.
- W2796869301 hasConcept C176217482 @default.
- W2796869301 hasConcept C187736073 @default.
- W2796869301 hasConcept C204321447 @default.
- W2796869301 hasConcept C21547014 @default.
- W2796869301 hasConcept C23123220 @default.
- W2796869301 hasConcept C2777530160 @default.
- W2796869301 hasConcept C2778739407 @default.
- W2796869301 hasConcept C2780451532 @default.
- W2796869301 hasConcept C33923547 @default.
- W2796869301 hasConcept C41008148 @default.
- W2796869301 hasConcept C41895202 @default.
- W2796869301 hasConcept C90805587 @default.
- W2796869301 hasConceptScore W2796869301C103278499 @default.
- W2796869301 hasConceptScore W2796869301C105795698 @default.
- W2796869301 hasConceptScore W2796869301C115961682 @default.
- W2796869301 hasConceptScore W2796869301C138885662 @default.
- W2796869301 hasConceptScore W2796869301C154945302 @default.
- W2796869301 hasConceptScore W2796869301C157657479 @default.
- W2796869301 hasConceptScore W2796869301C162324750 @default.
- W2796869301 hasConceptScore W2796869301C165064840 @default.
- W2796869301 hasConceptScore W2796869301C176217482 @default.