Matches in SemOpenAlex for { <https://semopenalex.org/work/W4289315680> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W4289315680 endingPage "109571" @default.
- W4289315680 startingPage "109571" @default.
- W4289315680 abstract "The deployment of deep convolutional neural networks (CNNs) is heavily constrained by its high computational costs and parameter redundancy. For this reason, general group convolution (GGC) and depthwise convolution (DWC) were proposed, but they limited the information transfer in the channel dimension. In this paper, a novel and efficient overlapped group convolution (OGC) is proposed to improve the information transfer between channels. In OGC, the input feature maps can be overlapped between different groups. Compared with GGC, OGC has better information transfer in the channel dimension without additional parameters and computational cost. In theory, OGC unifies the standard convolution (SDC), GGC, and DWC. In other words, SDC, GGC, and DWC all belong to the special cases of OGC. In OGC, two flexible hyperparameters are defined, the number of input feature maps in each group (g) and the stride between adjacent groups (s), which make OGC more flexible and can make the trade-off between accuracy and parameters. The performance of OGC is analyzed in terms of parameters, computational cost, accuracy, run time, etc. The classification and object detection tasks are used to evaluate the performance of OGC. Experimental results show that the OGC has higher accuracy and is more efficient than the corresponding SDC, GGC, and DWC. The ratio of the two hyperparameters in OGC has a great impact on accuracy. When 23<sg<67, OGC has higher accuracy than others. The proposed OGC is more stable and robust than GGC." @default.
- W4289315680 created "2022-08-02" @default.
- W4289315680 creator A5028288572 @default.
- W4289315680 creator A5071878141 @default.
- W4289315680 creator A5074140059 @default.
- W4289315680 creator A5081611476 @default.
- W4289315680 date "2022-10-01" @default.
- W4289315680 modified "2023-09-29" @default.
- W4289315680 title "OGCNet: Overlapped group convolution for deep convolutional neural networks" @default.
- W4289315680 cites W2097117768 @default.
- W4289315680 cites W2117539524 @default.
- W4289315680 cites W2183341477 @default.
- W4289315680 cites W2194775991 @default.
- W4289315680 cites W2302255633 @default.
- W4289315680 cites W2531409750 @default.
- W4289315680 cites W2549139847 @default.
- W4289315680 cites W2883780447 @default.
- W4289315680 cites W2898910301 @default.
- W4289315680 cites W2901950140 @default.
- W4289315680 cites W2910511383 @default.
- W4289315680 cites W2963125010 @default.
- W4289315680 cites W2963163009 @default.
- W4289315680 cites W2963993763 @default.
- W4289315680 cites W2964081807 @default.
- W4289315680 cites W2964350391 @default.
- W4289315680 cites W2965862350 @default.
- W4289315680 cites W2983108667 @default.
- W4289315680 cites W3002750555 @default.
- W4289315680 cites W3033835243 @default.
- W4289315680 cites W3047011367 @default.
- W4289315680 cites W3072011699 @default.
- W4289315680 cites W3095523882 @default.
- W4289315680 cites W3102564565 @default.
- W4289315680 cites W3134090706 @default.
- W4289315680 cites W3155497642 @default.
- W4289315680 cites W3170968300 @default.
- W4289315680 cites W3190215629 @default.
- W4289315680 cites W3199986538 @default.
- W4289315680 cites W3205764667 @default.
- W4289315680 cites W4200122786 @default.
- W4289315680 cites W4214604703 @default.
- W4289315680 cites W639708223 @default.
- W4289315680 doi "https://doi.org/10.1016/j.knosys.2022.109571" @default.
- W4289315680 hasPublicationYear "2022" @default.
- W4289315680 type Work @default.
- W4289315680 citedByCount "1" @default.
- W4289315680 countsByYear W42893156802023 @default.
- W4289315680 crossrefType "journal-article" @default.
- W4289315680 hasAuthorship W4289315680A5028288572 @default.
- W4289315680 hasAuthorship W4289315680A5071878141 @default.
- W4289315680 hasAuthorship W4289315680A5074140059 @default.
- W4289315680 hasAuthorship W4289315680A5081611476 @default.
- W4289315680 hasConcept C153180895 @default.
- W4289315680 hasConcept C154945302 @default.
- W4289315680 hasConcept C202444582 @default.
- W4289315680 hasConcept C33676613 @default.
- W4289315680 hasConcept C33923547 @default.
- W4289315680 hasConcept C41008148 @default.
- W4289315680 hasConcept C45347329 @default.
- W4289315680 hasConcept C50644808 @default.
- W4289315680 hasConcept C81363708 @default.
- W4289315680 hasConcept C8642999 @default.
- W4289315680 hasConceptScore W4289315680C153180895 @default.
- W4289315680 hasConceptScore W4289315680C154945302 @default.
- W4289315680 hasConceptScore W4289315680C202444582 @default.
- W4289315680 hasConceptScore W4289315680C33676613 @default.
- W4289315680 hasConceptScore W4289315680C33923547 @default.
- W4289315680 hasConceptScore W4289315680C41008148 @default.
- W4289315680 hasConceptScore W4289315680C45347329 @default.
- W4289315680 hasConceptScore W4289315680C50644808 @default.
- W4289315680 hasConceptScore W4289315680C81363708 @default.
- W4289315680 hasConceptScore W4289315680C8642999 @default.
- W4289315680 hasLocation W42893156801 @default.
- W4289315680 hasOpenAccess W4289315680 @default.
- W4289315680 hasPrimaryLocation W42893156801 @default.
- W4289315680 hasRelatedWork W2295021132 @default.
- W4289315680 hasRelatedWork W2755240195 @default.
- W4289315680 hasRelatedWork W2766634277 @default.
- W4289315680 hasRelatedWork W2767651786 @default.
- W4289315680 hasRelatedWork W2912288872 @default.
- W4289315680 hasRelatedWork W3130227562 @default.
- W4289315680 hasRelatedWork W4206951940 @default.
- W4289315680 hasRelatedWork W4304182771 @default.
- W4289315680 hasRelatedWork W4312417841 @default.
- W4289315680 hasRelatedWork W564581980 @default.
- W4289315680 hasVolume "253" @default.
- W4289315680 isParatext "false" @default.
- W4289315680 isRetracted "false" @default.
- W4289315680 workType "article" @default.