Matches in SemOpenAlex for { <https://semopenalex.org/work/W2949649078> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W2949649078 abstract "Convolutional neural networks (CNNs) have been widely deployed in the fields of computer vision and pattern recognition because of their high accuracy. However, large convolution operations are computing-intensive that often requires a powerful computing platform such as Graphics Processing Unit (GPU). This makes it difficult to apply CNNs to portable devices. The state-of-the-art CNNs, such as MobileNetV2 and Xception, adopt depthwise separable convolution to replace the standard convolution for embedded platforms. That significantly reduces operations and parameters with only limited loss in accuracy. This highly structured model is very suitable for Field-Programmable Gate Array (FPGA) implementation. In this paper, a scalable high performance depthwise separable convolution optimized CNN accelerator is proposed. The accelerator can be fit into an FPGA of different sizes, provided the balancing between hardware resources and processing speed. As an example, MobileNetV2 is implemented on Arria 10 SoC FPGA, and the results show this accelerator can classify each picture from ImageNet in 3.75ms, which is about 266.6 frames per second. This achieves 20x speedup if compared to CPU." @default.
- W2949649078 created "2019-06-27" @default.
- W2949649078 creator A5042654091 @default.
- W2949649078 creator A5056594372 @default.
- W2949649078 creator A5058293888 @default.
- W2949649078 date "2018-09-03" @default.
- W2949649078 modified "2023-09-23" @default.
- W2949649078 title "A CNN Accelerator on FPGA Using Depthwise Separable Convolution" @default.
- W2949649078 cites W1549358575 @default.
- W2949649078 cites W1594170634 @default.
- W2949649078 cites W1825672851 @default.
- W2949649078 cites W1841592590 @default.
- W2949649078 cites W2048266589 @default.
- W2949649078 cites W2067523571 @default.
- W2949649078 cites W2276486856 @default.
- W2949649078 cites W2289252105 @default.
- W2949649078 cites W2523838129 @default.
- W2949649078 cites W2525740295 @default.
- W2949649078 cites W2527351224 @default.
- W2949649078 cites W2584311934 @default.
- W2949649078 cites W2612445135 @default.
- W2949649078 cites W2623629680 @default.
- W2949649078 cites W2627042741 @default.
- W2949649078 cites W2783000019 @default.
- W2949649078 cites W2789333991 @default.
- W2949649078 cites W2796438033 @default.
- W2949649078 cites W2949117887 @default.
- W2949649078 cites W2951583185 @default.
- W2949649078 doi "https://doi.org/10.48550/arxiv.1809.01536" @default.
- W2949649078 hasPublicationYear "2018" @default.
- W2949649078 type Work @default.
- W2949649078 sameAs 2949649078 @default.
- W2949649078 citedByCount "0" @default.
- W2949649078 crossrefType "posted-content" @default.
- W2949649078 hasAuthorship W2949649078A5042654091 @default.
- W2949649078 hasAuthorship W2949649078A5056594372 @default.
- W2949649078 hasAuthorship W2949649078A5058293888 @default.
- W2949649078 hasBestOaLocation W29496490781 @default.
- W2949649078 hasConcept C114614502 @default.
- W2949649078 hasConcept C13164978 @default.
- W2949649078 hasConcept C134306372 @default.
- W2949649078 hasConcept C154945302 @default.
- W2949649078 hasConcept C173608175 @default.
- W2949649078 hasConcept C2779851693 @default.
- W2949649078 hasConcept C33923547 @default.
- W2949649078 hasConcept C41008148 @default.
- W2949649078 hasConcept C42935608 @default.
- W2949649078 hasConcept C45347329 @default.
- W2949649078 hasConcept C459310 @default.
- W2949649078 hasConcept C48044578 @default.
- W2949649078 hasConcept C50644808 @default.
- W2949649078 hasConcept C68339613 @default.
- W2949649078 hasConcept C70710897 @default.
- W2949649078 hasConcept C74193536 @default.
- W2949649078 hasConcept C77088390 @default.
- W2949649078 hasConcept C81363708 @default.
- W2949649078 hasConcept C9390403 @default.
- W2949649078 hasConceptScore W2949649078C114614502 @default.
- W2949649078 hasConceptScore W2949649078C13164978 @default.
- W2949649078 hasConceptScore W2949649078C134306372 @default.
- W2949649078 hasConceptScore W2949649078C154945302 @default.
- W2949649078 hasConceptScore W2949649078C173608175 @default.
- W2949649078 hasConceptScore W2949649078C2779851693 @default.
- W2949649078 hasConceptScore W2949649078C33923547 @default.
- W2949649078 hasConceptScore W2949649078C41008148 @default.
- W2949649078 hasConceptScore W2949649078C42935608 @default.
- W2949649078 hasConceptScore W2949649078C45347329 @default.
- W2949649078 hasConceptScore W2949649078C459310 @default.
- W2949649078 hasConceptScore W2949649078C48044578 @default.
- W2949649078 hasConceptScore W2949649078C50644808 @default.
- W2949649078 hasConceptScore W2949649078C68339613 @default.
- W2949649078 hasConceptScore W2949649078C70710897 @default.
- W2949649078 hasConceptScore W2949649078C74193536 @default.
- W2949649078 hasConceptScore W2949649078C77088390 @default.
- W2949649078 hasConceptScore W2949649078C81363708 @default.
- W2949649078 hasConceptScore W2949649078C9390403 @default.
- W2949649078 hasLocation W29496490781 @default.
- W2949649078 hasOpenAccess W2949649078 @default.
- W2949649078 hasPrimaryLocation W29496490781 @default.
- W2949649078 hasRelatedWork W2048593763 @default.
- W2949649078 hasRelatedWork W2100501042 @default.
- W2949649078 hasRelatedWork W2116951845 @default.
- W2949649078 hasRelatedWork W2370911386 @default.
- W2949649078 hasRelatedWork W2565158842 @default.
- W2949649078 hasRelatedWork W2770717529 @default.
- W2949649078 hasRelatedWork W2794923745 @default.
- W2949649078 hasRelatedWork W2949649078 @default.
- W2949649078 hasRelatedWork W3015861875 @default.
- W2949649078 hasRelatedWork W3179800311 @default.
- W2949649078 isParatext "false" @default.
- W2949649078 isRetracted "false" @default.
- W2949649078 magId "2949649078" @default.
- W2949649078 workType "article" @default.