Matches in SemOpenAlex for { <https://semopenalex.org/work/W3080521489> ?p ?o ?g. }
- W3080521489 abstract "Standard video and movie description tasks abstract away from person identities, thus failing to link identities across sentences. We propose a multi-sentence Identity-Aware Video Description task, which overcomes this limitation and requires to re-identify persons locally within a set of consecutive clips. We introduce an auxiliary task of Fill-in the Identity, that aims to predict persons' IDs consistently within a set of clips, when the video descriptions are given. Our proposed approach to this task leverages a Transformer architecture allowing for coherent joint prediction of multiple IDs. One of the key components is a gender-aware textual representation as well an additional gender prediction objective in the main model. This auxiliary task allows us to propose a two-stage approach to Identity-Aware Video Description. We first generate multi-sentence video descriptions, and then apply our Fill-in the Identity model to establish links between the predicted person entities. To be able to tackle both tasks, we augment the Large Scale Movie Description Challenge (LSMDC) benchmark with new annotations suited for our problem statement. Experiments show that our proposed Fill-in the Identity model is superior to several baselines and recent works, and allows us to generate descriptions with locally re-identified people." @default.
- W3080521489 created "2020-09-01" @default.
- W3080521489 creator A5029105520 @default.
- W3080521489 creator A5030680279 @default.
- W3080521489 creator A5037747070 @default.
- W3080521489 date "2020-08-22" @default.
- W3080521489 modified "2023-09-27" @default.
- W3080521489 title "Identity-Aware Multi-Sentence Video Description" @default.
- W3080521489 cites W1572567476 @default.
- W3080521489 cites W1586939924 @default.
- W3080521489 cites W1596841185 @default.
- W3080521489 cites W1673310716 @default.
- W3080521489 cites W1893116441 @default.
- W3080521489 cites W1947481528 @default.
- W3080521489 cites W1956340063 @default.
- W3080521489 cites W2055251102 @default.
- W3080521489 cites W2101105183 @default.
- W3080521489 cites W2108598243 @default.
- W3080521489 cites W2119031011 @default.
- W3080521489 cites W2121027212 @default.
- W3080521489 cites W2133459682 @default.
- W3080521489 cites W2139501017 @default.
- W3080521489 cites W2144767994 @default.
- W3080521489 cites W2145287260 @default.
- W3080521489 cites W2168996682 @default.
- W3080521489 cites W2194775991 @default.
- W3080521489 cites W2325939864 @default.
- W3080521489 cites W2507009361 @default.
- W3080521489 cites W2556388456 @default.
- W3080521489 cites W2562836854 @default.
- W3080521489 cites W2565656701 @default.
- W3080521489 cites W2605585413 @default.
- W3080521489 cites W2619947201 @default.
- W3080521489 cites W2798725893 @default.
- W3080521489 cites W2798793675 @default.
- W3080521489 cites W2883910824 @default.
- W3080521489 cites W2891939431 @default.
- W3080521489 cites W2914699769 @default.
- W3080521489 cites W2949365443 @default.
- W3080521489 cites W2951183276 @default.
- W3080521489 cites W2962698660 @default.
- W3080521489 cites W2962799512 @default.
- W3080521489 cites W2962937869 @default.
- W3080521489 cites W2963177403 @default.
- W3080521489 cites W2963341956 @default.
- W3080521489 cites W2963351113 @default.
- W3080521489 cites W2963403868 @default.
- W3080521489 cites W2963498278 @default.
- W3080521489 cites W2963524571 @default.
- W3080521489 cites W2963552819 @default.
- W3080521489 cites W2963576560 @default.
- W3080521489 cites W2963753226 @default.
- W3080521489 cites W2963811641 @default.
- W3080521489 cites W2963839617 @default.
- W3080521489 cites W2963843052 @default.
- W3080521489 cites W2963916161 @default.
- W3080521489 cites W2964102650 @default.
- W3080521489 cites W2968101724 @default.
- W3080521489 cites W2984862483 @default.
- W3080521489 cites W2985144848 @default.
- W3080521489 cites W2988753485 @default.
- W3080521489 cites W2989322838 @default.
- W3080521489 cites W3098682680 @default.
- W3080521489 cites W3101227480 @default.
- W3080521489 cites W3101998545 @default.
- W3080521489 cites W38568571 @default.
- W3080521489 hasPublicationYear "2020" @default.
- W3080521489 type Work @default.
- W3080521489 sameAs 3080521489 @default.
- W3080521489 citedByCount "0" @default.
- W3080521489 crossrefType "posted-content" @default.
- W3080521489 hasAuthorship W3080521489A5029105520 @default.
- W3080521489 hasAuthorship W3080521489A5030680279 @default.
- W3080521489 hasAuthorship W3080521489A5037747070 @default.
- W3080521489 hasConcept C107457646 @default.
- W3080521489 hasConcept C121332964 @default.
- W3080521489 hasConcept C13280743 @default.
- W3080521489 hasConcept C138885662 @default.
- W3080521489 hasConcept C154945302 @default.
- W3080521489 hasConcept C162324750 @default.
- W3080521489 hasConcept C165801399 @default.
- W3080521489 hasConcept C177264268 @default.
- W3080521489 hasConcept C17744445 @default.
- W3080521489 hasConcept C185798385 @default.
- W3080521489 hasConcept C187736073 @default.
- W3080521489 hasConcept C199360897 @default.
- W3080521489 hasConcept C199539241 @default.
- W3080521489 hasConcept C204321447 @default.
- W3080521489 hasConcept C205649164 @default.
- W3080521489 hasConcept C23123220 @default.
- W3080521489 hasConcept C24890656 @default.
- W3080521489 hasConcept C2776359362 @default.
- W3080521489 hasConcept C2777026412 @default.
- W3080521489 hasConcept C2777530160 @default.
- W3080521489 hasConcept C2778355321 @default.
- W3080521489 hasConcept C2780451532 @default.
- W3080521489 hasConcept C41008148 @default.
- W3080521489 hasConcept C41895202 @default.
- W3080521489 hasConcept C62520636 @default.
- W3080521489 hasConcept C66322947 @default.