Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313446052> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W4313446052 abstract "The ability to distinguish between different movie scenes is critical for understanding the storyline of a movie. However, accurately detecting movie scenes is often challenging as it requires the ability to reason over very long movie segments. This is in contrast to most existing video recognition models, which are typically designed for short-range video analysis. This work proposes a State-Space Transformer model that can efficiently capture dependencies in long movie videos for accurate movie scene detection. Our model, dubbed TranS4mer, is built using a novel S4A building block, which combines the strengths of structured state-space sequence (S4) and self-attention (A) layers. Given a sequence of frames divided into movie shots (uninterrupted periods where the camera position does not change), the S4A block first applies self-attention to capture short-range intra-shot dependencies. Afterward, the state-space operation in the S4A block is used to aggregate long-range inter-shot cues. The final TranS4mer model, which can be trained end-to-end, is obtained by stacking the S4A blocks one after the other multiple times. Our proposed TranS4mer outperforms all prior methods in three movie scene detection datasets, including MovieNet, BBC, and OVSD, while also being $2times$ faster and requiring $3times$ less GPU memory than standard Transformer models. We will release our code and models." @default.
- W4313446052 created "2023-01-06" @default.
- W4313446052 creator A5063715282 @default.
- W4313446052 creator A5073282870 @default.
- W4313446052 creator A5081800468 @default.
- W4313446052 creator A5084606101 @default.
- W4313446052 creator A5090463221 @default.
- W4313446052 date "2022-12-29" @default.
- W4313446052 modified "2023-09-26" @default.
- W4313446052 title "Efficient Movie Scene Detection using State-Space Transformers" @default.
- W4313446052 doi "https://doi.org/10.48550/arxiv.2212.14427" @default.
- W4313446052 hasPublicationYear "2022" @default.
- W4313446052 type Work @default.
- W4313446052 citedByCount "0" @default.
- W4313446052 crossrefType "posted-content" @default.
- W4313446052 hasAuthorship W4313446052A5063715282 @default.
- W4313446052 hasAuthorship W4313446052A5073282870 @default.
- W4313446052 hasAuthorship W4313446052A5081800468 @default.
- W4313446052 hasAuthorship W4313446052A5084606101 @default.
- W4313446052 hasAuthorship W4313446052A5090463221 @default.
- W4313446052 hasBestOaLocation W43134460521 @default.
- W4313446052 hasConcept C119599485 @default.
- W4313446052 hasConcept C121684516 @default.
- W4313446052 hasConcept C127413603 @default.
- W4313446052 hasConcept C154945302 @default.
- W4313446052 hasConcept C165801399 @default.
- W4313446052 hasConcept C178790620 @default.
- W4313446052 hasConcept C185592680 @default.
- W4313446052 hasConcept C2524010 @default.
- W4313446052 hasConcept C2777210771 @default.
- W4313446052 hasConcept C2778112365 @default.
- W4313446052 hasConcept C2778344882 @default.
- W4313446052 hasConcept C31972630 @default.
- W4313446052 hasConcept C33923547 @default.
- W4313446052 hasConcept C41008148 @default.
- W4313446052 hasConcept C54355233 @default.
- W4313446052 hasConcept C66322947 @default.
- W4313446052 hasConcept C86803240 @default.
- W4313446052 hasConceptScore W4313446052C119599485 @default.
- W4313446052 hasConceptScore W4313446052C121684516 @default.
- W4313446052 hasConceptScore W4313446052C127413603 @default.
- W4313446052 hasConceptScore W4313446052C154945302 @default.
- W4313446052 hasConceptScore W4313446052C165801399 @default.
- W4313446052 hasConceptScore W4313446052C178790620 @default.
- W4313446052 hasConceptScore W4313446052C185592680 @default.
- W4313446052 hasConceptScore W4313446052C2524010 @default.
- W4313446052 hasConceptScore W4313446052C2777210771 @default.
- W4313446052 hasConceptScore W4313446052C2778112365 @default.
- W4313446052 hasConceptScore W4313446052C2778344882 @default.
- W4313446052 hasConceptScore W4313446052C31972630 @default.
- W4313446052 hasConceptScore W4313446052C33923547 @default.
- W4313446052 hasConceptScore W4313446052C41008148 @default.
- W4313446052 hasConceptScore W4313446052C54355233 @default.
- W4313446052 hasConceptScore W4313446052C66322947 @default.
- W4313446052 hasConceptScore W4313446052C86803240 @default.
- W4313446052 hasLocation W43134460521 @default.
- W4313446052 hasOpenAccess W4313446052 @default.
- W4313446052 hasPrimaryLocation W43134460521 @default.
- W4313446052 hasRelatedWork W1592626709 @default.
- W4313446052 hasRelatedWork W1891287906 @default.
- W4313446052 hasRelatedWork W1969923398 @default.
- W4313446052 hasRelatedWork W2036807459 @default.
- W4313446052 hasRelatedWork W2056180080 @default.
- W4313446052 hasRelatedWork W2104432455 @default.
- W4313446052 hasRelatedWork W2166044122 @default.
- W4313446052 hasRelatedWork W2387675639 @default.
- W4313446052 hasRelatedWork W2397188463 @default.
- W4313446052 hasRelatedWork W2747873377 @default.
- W4313446052 isParatext "false" @default.
- W4313446052 isRetracted "false" @default.
- W4313446052 workType "article" @default.