Matches in SemOpenAlex for { <https://semopenalex.org/work/W3112702248> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W3112702248 abstract "Deep learning algorithms have demonstrated remarkable performance in many sectors and have become one of the main foundations of modern computer-vision solutions. However, these algorithms often impose prohibitive levels of memory and computational overhead, especially in resource-constrained environments. In this study, we combine the state-of-the-art object-detection model YOLOv3 with depthwise separable convolutions and variational dropout in an attempt to bridge the gap between the superior accuracy of convolutional neural networks and the limited access to computational resources. We propose three lightweight variants of YOLOv3 by replacing the original network’s standard convolutions with depthwise separable convolutions at different strategic locations within the network, and we evaluate their impacts on YOLOv3’s size, speed, and accuracy. We also explore variational dropout: a technique that finds individual and unbounded dropout rates for each neural network weight. Experiments on the PASCAL VOC benchmark dataset show promising results where variational dropout combined with the most efficient YOLOv3 variant lead to an extremely sparse solution that reduces 95% of the baseline network’s parameters at a relatively small drop of 3% in accuracy." @default.
- W3112702248 created "2020-12-21" @default.
- W3112702248 creator A5068284344 @default.
- W3112702248 creator A5084446270 @default.
- W3112702248 creator A5088417940 @default.
- W3112702248 date "2020-01-01" @default.
- W3112702248 modified "2023-10-07" @default.
- W3112702248 title "Depthwise Separable Convolutions and Variational Dropout within the context of YOLOv3" @default.
- W3112702248 cites W1436538595 @default.
- W3112702248 cites W1536680647 @default.
- W3112702248 cites W1861492603 @default.
- W3112702248 cites W2031489346 @default.
- W3112702248 cites W2102605133 @default.
- W3112702248 cites W2117539524 @default.
- W3112702248 cites W2570343428 @default.
- W3112702248 cites W2618530766 @default.
- W3112702248 cites W2796114583 @default.
- W3112702248 cites W2900662096 @default.
- W3112702248 cites W2913994632 @default.
- W3112702248 cites W2962976294 @default.
- W3112702248 cites W2963037989 @default.
- W3112702248 cites W2963346150 @default.
- W3112702248 cites W2972309292 @default.
- W3112702248 cites W2973696262 @default.
- W3112702248 cites W2996575742 @default.
- W3112702248 cites W2999032021 @default.
- W3112702248 cites W639708223 @default.
- W3112702248 doi "https://doi.org/10.1007/978-3-030-64556-4_9" @default.
- W3112702248 hasPublicationYear "2020" @default.
- W3112702248 type Work @default.
- W3112702248 sameAs 3112702248 @default.
- W3112702248 citedByCount "1" @default.
- W3112702248 countsByYear W31127022482022 @default.
- W3112702248 crossrefType "book-chapter" @default.
- W3112702248 hasAuthorship W3112702248A5068284344 @default.
- W3112702248 hasAuthorship W3112702248A5084446270 @default.
- W3112702248 hasAuthorship W3112702248A5088417940 @default.
- W3112702248 hasConcept C11413529 @default.
- W3112702248 hasConcept C119857082 @default.
- W3112702248 hasConcept C126255220 @default.
- W3112702248 hasConcept C13280743 @default.
- W3112702248 hasConcept C134306372 @default.
- W3112702248 hasConcept C151730666 @default.
- W3112702248 hasConcept C154945302 @default.
- W3112702248 hasConcept C185798385 @default.
- W3112702248 hasConcept C199360897 @default.
- W3112702248 hasConcept C205649164 @default.
- W3112702248 hasConcept C2776145597 @default.
- W3112702248 hasConcept C2779343474 @default.
- W3112702248 hasConcept C33923547 @default.
- W3112702248 hasConcept C41008148 @default.
- W3112702248 hasConcept C50644808 @default.
- W3112702248 hasConcept C70710897 @default.
- W3112702248 hasConcept C75608658 @default.
- W3112702248 hasConcept C80444323 @default.
- W3112702248 hasConcept C86803240 @default.
- W3112702248 hasConceptScore W3112702248C11413529 @default.
- W3112702248 hasConceptScore W3112702248C119857082 @default.
- W3112702248 hasConceptScore W3112702248C126255220 @default.
- W3112702248 hasConceptScore W3112702248C13280743 @default.
- W3112702248 hasConceptScore W3112702248C134306372 @default.
- W3112702248 hasConceptScore W3112702248C151730666 @default.
- W3112702248 hasConceptScore W3112702248C154945302 @default.
- W3112702248 hasConceptScore W3112702248C185798385 @default.
- W3112702248 hasConceptScore W3112702248C199360897 @default.
- W3112702248 hasConceptScore W3112702248C205649164 @default.
- W3112702248 hasConceptScore W3112702248C2776145597 @default.
- W3112702248 hasConceptScore W3112702248C2779343474 @default.
- W3112702248 hasConceptScore W3112702248C33923547 @default.
- W3112702248 hasConceptScore W3112702248C41008148 @default.
- W3112702248 hasConceptScore W3112702248C50644808 @default.
- W3112702248 hasConceptScore W3112702248C70710897 @default.
- W3112702248 hasConceptScore W3112702248C75608658 @default.
- W3112702248 hasConceptScore W3112702248C80444323 @default.
- W3112702248 hasConceptScore W3112702248C86803240 @default.
- W3112702248 hasLocation W31127022481 @default.
- W3112702248 hasOpenAccess W3112702248 @default.
- W3112702248 hasPrimaryLocation W31127022481 @default.
- W3112702248 hasRelatedWork W13704299 @default.
- W3112702248 hasRelatedWork W2433769 @default.
- W3112702248 hasRelatedWork W2701911 @default.
- W3112702248 hasRelatedWork W3540334 @default.
- W3112702248 hasRelatedWork W378023 @default.
- W3112702248 hasRelatedWork W4919037 @default.
- W3112702248 hasRelatedWork W4964196 @default.
- W3112702248 hasRelatedWork W5939791 @default.
- W3112702248 hasRelatedWork W8772801 @default.
- W3112702248 hasRelatedWork W9770290 @default.
- W3112702248 isParatext "false" @default.
- W3112702248 isRetracted "false" @default.
- W3112702248 magId "3112702248" @default.
- W3112702248 workType "book-chapter" @default.