Matches in SemOpenAlex for { <https://semopenalex.org/work/W4372342010> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W4372342010 abstract "High-performance video frame interpolation is challenging for complex scenes with diverse motion and occlusion characteristics. Existing methods, deploying off-the-shelf flow estimators to acquire initial characterizations refined by multiple subsequent models, often require heavy network architectures that are not practical for resource constrained systems. We investigate the unary potentials of the characterizations to improve efficiency. Specifically, we design a lightweight neural network to construct motion volumes via ensembles of offset approximations, and propose a spatiotemporal variance-aware loss to supervise the network learning. For network compactness, our spatiotemporal variance-supervised motion volume (SVMV) utilizes shared spatiotemporal representations via correlations among approximations, of which the diversifications are exploited to better leverage the network’s expressiveness through the spatiotemporal variances of motions and occlusions within the time interval to be interpolated. Experiments on publicly available datasets show that our method performs favorably against existing methods with a more compact network and less runtime." @default.
- W4372342010 created "2023-05-07" @default.
- W4372342010 creator A5004164569 @default.
- W4372342010 creator A5035112538 @default.
- W4372342010 creator A5086595168 @default.
- W4372342010 date "2023-06-04" @default.
- W4372342010 modified "2023-09-27" @default.
- W4372342010 title "SVMV: Spatiotemporal Variance-Supervised Motion Volume for Video Frame Interpolation" @default.
- W4372342010 cites W2470139095 @default.
- W4372342010 cites W2738579427 @default.
- W4372342010 cites W2949258649 @default.
- W4372342010 cites W2963782415 @default.
- W4372342010 cites W2964251418 @default.
- W4372342010 cites W2973673960 @default.
- W4372342010 cites W2985775862 @default.
- W4372342010 cites W2997150500 @default.
- W4372342010 cites W3034921716 @default.
- W4372342010 cites W3035239272 @default.
- W4372342010 cites W3035552210 @default.
- W4372342010 cites W3083550621 @default.
- W4372342010 cites W3102015846 @default.
- W4372342010 cites W3128053054 @default.
- W4372342010 cites W3184828860 @default.
- W4372342010 cites W3203041407 @default.
- W4372342010 cites W4224933792 @default.
- W4372342010 cites W4312324876 @default.
- W4372342010 cites W4312364368 @default.
- W4372342010 doi "https://doi.org/10.1109/icassp49357.2023.10094764" @default.
- W4372342010 hasPublicationYear "2023" @default.
- W4372342010 type Work @default.
- W4372342010 citedByCount "0" @default.
- W4372342010 crossrefType "proceedings-article" @default.
- W4372342010 hasAuthorship W4372342010A5004164569 @default.
- W4372342010 hasAuthorship W4372342010A5035112538 @default.
- W4372342010 hasAuthorship W4372342010A5086595168 @default.
- W4372342010 hasBestOaLocation W43723420101 @default.
- W4372342010 hasConcept C104114177 @default.
- W4372342010 hasConcept C105795698 @default.
- W4372342010 hasConcept C11413529 @default.
- W4372342010 hasConcept C121955636 @default.
- W4372342010 hasConcept C13280743 @default.
- W4372342010 hasConcept C137800194 @default.
- W4372342010 hasConcept C144133560 @default.
- W4372342010 hasConcept C153083717 @default.
- W4372342010 hasConcept C154945302 @default.
- W4372342010 hasConcept C175291020 @default.
- W4372342010 hasConcept C176809094 @default.
- W4372342010 hasConcept C185429906 @default.
- W4372342010 hasConcept C196083921 @default.
- W4372342010 hasConcept C199360897 @default.
- W4372342010 hasConcept C205649164 @default.
- W4372342010 hasConcept C31972630 @default.
- W4372342010 hasConcept C33923547 @default.
- W4372342010 hasConcept C41008148 @default.
- W4372342010 hasConceptScore W4372342010C104114177 @default.
- W4372342010 hasConceptScore W4372342010C105795698 @default.
- W4372342010 hasConceptScore W4372342010C11413529 @default.
- W4372342010 hasConceptScore W4372342010C121955636 @default.
- W4372342010 hasConceptScore W4372342010C13280743 @default.
- W4372342010 hasConceptScore W4372342010C137800194 @default.
- W4372342010 hasConceptScore W4372342010C144133560 @default.
- W4372342010 hasConceptScore W4372342010C153083717 @default.
- W4372342010 hasConceptScore W4372342010C154945302 @default.
- W4372342010 hasConceptScore W4372342010C175291020 @default.
- W4372342010 hasConceptScore W4372342010C176809094 @default.
- W4372342010 hasConceptScore W4372342010C185429906 @default.
- W4372342010 hasConceptScore W4372342010C196083921 @default.
- W4372342010 hasConceptScore W4372342010C199360897 @default.
- W4372342010 hasConceptScore W4372342010C205649164 @default.
- W4372342010 hasConceptScore W4372342010C31972630 @default.
- W4372342010 hasConceptScore W4372342010C33923547 @default.
- W4372342010 hasConceptScore W4372342010C41008148 @default.
- W4372342010 hasLocation W43723420101 @default.
- W4372342010 hasOpenAccess W4372342010 @default.
- W4372342010 hasPrimaryLocation W43723420101 @default.
- W4372342010 hasRelatedWork W2029249305 @default.
- W4372342010 hasRelatedWork W2098025439 @default.
- W4372342010 hasRelatedWork W2115571026 @default.
- W4372342010 hasRelatedWork W2511137960 @default.
- W4372342010 hasRelatedWork W2604231787 @default.
- W4372342010 hasRelatedWork W2610014769 @default.
- W4372342010 hasRelatedWork W2687972263 @default.
- W4372342010 hasRelatedWork W2897576080 @default.
- W4372342010 hasRelatedWork W3214088465 @default.
- W4372342010 hasRelatedWork W4321142835 @default.
- W4372342010 isParatext "false" @default.
- W4372342010 isRetracted "false" @default.
- W4372342010 workType "article" @default.