Matches in SemOpenAlex for { <https://semopenalex.org/work/W3013559402> ?p ?o ?g. }
- W3013559402 endingPage "59759" @default.
- W3013559402 startingPage "59750" @default.
- W3013559402 abstract "Convolutional neural network (CNN) is a powerful tool for many data applications. However, its high dimension nature, large network size and computational complexity, and the need of large amount of training data make it challenging to be used in edge computing applications, which are becoming increasingly popular, relevant and important. In this paper, we propose a descriptor based approach to accelerate convolutional neural network training, reduce input dimension and network size, which greatly facilitates the use of CNN for edge computating and even cloud computing. By using image descriptors to extract features from original images, we report a simpler convolutional neural network with fast training speed, low memory usage and outstanding accuracy without the need for a pre-trained network as opposed to the state of art. In indoor localization, our SURF descriptors accelerated CNN (SurfCNN) can reach an average position error of 0.28 m and orientation error of 9.2°. Compared to the conventional CNN that uses original images as input, our algorithm reduces the dimension of the input features by a factor of 48 without impairing the accuracy. Further, at an extreme feature reduction of 14,440 times, our model still retains an average position error retained at 0.41 m and orientation error at 14°." @default.
- W3013559402 created "2020-04-03" @default.
- W3013559402 creator A5009705432 @default.
- W3013559402 creator A5049362473 @default.
- W3013559402 creator A5077415473 @default.
- W3013559402 creator A5085240055 @default.
- W3013559402 creator A5089289729 @default.
- W3013559402 date "2020-01-01" @default.
- W3013559402 modified "2023-10-18" @default.
- W3013559402 title "SurfCNN: A Descriptor Accelerated Convolutional Neural Network for Image-Based Indoor Localization" @default.
- W3013559402 cites W1869500417 @default.
- W3013559402 cites W1955055330 @default.
- W3013559402 cites W1979266466 @default.
- W3013559402 cites W1980564965 @default.
- W3013559402 cites W2021851106 @default.
- W3013559402 cites W2042819305 @default.
- W3013559402 cites W2058535340 @default.
- W3013559402 cites W2064675550 @default.
- W3013559402 cites W2080863493 @default.
- W3013559402 cites W2097117768 @default.
- W3013559402 cites W2100989187 @default.
- W3013559402 cites W2102481828 @default.
- W3013559402 cites W2117228865 @default.
- W3013559402 cites W2119605622 @default.
- W3013559402 cites W2125556102 @default.
- W3013559402 cites W2129000642 @default.
- W3013559402 cites W2151103935 @default.
- W3013559402 cites W2159200188 @default.
- W3013559402 cites W2161381512 @default.
- W3013559402 cites W2180507614 @default.
- W3013559402 cites W2183341477 @default.
- W3013559402 cites W2194775991 @default.
- W3013559402 cites W2200124539 @default.
- W3013559402 cites W2213387816 @default.
- W3013559402 cites W2461086877 @default.
- W3013559402 cites W2522940611 @default.
- W3013559402 cites W2557998141 @default.
- W3013559402 cites W2584731199 @default.
- W3013559402 cites W2602709638 @default.
- W3013559402 cites W2605111497 @default.
- W3013559402 cites W2620475885 @default.
- W3013559402 cites W2737630486 @default.
- W3013559402 cites W2749379418 @default.
- W3013559402 cites W2963024893 @default.
- W3013559402 cites W2964175348 @default.
- W3013559402 cites W3102327032 @default.
- W3013559402 cites W3103648783 @default.
- W3013559402 cites W3124420883 @default.
- W3013559402 cites W4298069009 @default.
- W3013559402 doi "https://doi.org/10.1109/access.2020.2981620" @default.
- W3013559402 hasPublicationYear "2020" @default.
- W3013559402 type Work @default.
- W3013559402 sameAs 3013559402 @default.
- W3013559402 citedByCount "5" @default.
- W3013559402 countsByYear W30135594022020 @default.
- W3013559402 countsByYear W30135594022021 @default.
- W3013559402 countsByYear W30135594022022 @default.
- W3013559402 crossrefType "journal-article" @default.
- W3013559402 hasAuthorship W3013559402A5009705432 @default.
- W3013559402 hasAuthorship W3013559402A5049362473 @default.
- W3013559402 hasAuthorship W3013559402A5077415473 @default.
- W3013559402 hasAuthorship W3013559402A5085240055 @default.
- W3013559402 hasAuthorship W3013559402A5089289729 @default.
- W3013559402 hasBestOaLocation W30135594021 @default.
- W3013559402 hasConcept C10138342 @default.
- W3013559402 hasConcept C108583219 @default.
- W3013559402 hasConcept C111335779 @default.
- W3013559402 hasConcept C111919701 @default.
- W3013559402 hasConcept C11413529 @default.
- W3013559402 hasConcept C138236772 @default.
- W3013559402 hasConcept C138885662 @default.
- W3013559402 hasConcept C153180895 @default.
- W3013559402 hasConcept C154945302 @default.
- W3013559402 hasConcept C162307627 @default.
- W3013559402 hasConcept C162324750 @default.
- W3013559402 hasConcept C16345878 @default.
- W3013559402 hasConcept C179799912 @default.
- W3013559402 hasConcept C198082294 @default.
- W3013559402 hasConcept C202444582 @default.
- W3013559402 hasConcept C2524010 @default.
- W3013559402 hasConcept C2776401178 @default.
- W3013559402 hasConcept C31972630 @default.
- W3013559402 hasConcept C33676613 @default.
- W3013559402 hasConcept C33923547 @default.
- W3013559402 hasConcept C41008148 @default.
- W3013559402 hasConcept C41895202 @default.
- W3013559402 hasConcept C50644808 @default.
- W3013559402 hasConcept C79974875 @default.
- W3013559402 hasConcept C81363708 @default.
- W3013559402 hasConceptScore W3013559402C10138342 @default.
- W3013559402 hasConceptScore W3013559402C108583219 @default.
- W3013559402 hasConceptScore W3013559402C111335779 @default.
- W3013559402 hasConceptScore W3013559402C111919701 @default.
- W3013559402 hasConceptScore W3013559402C11413529 @default.
- W3013559402 hasConceptScore W3013559402C138236772 @default.
- W3013559402 hasConceptScore W3013559402C138885662 @default.
- W3013559402 hasConceptScore W3013559402C153180895 @default.
- W3013559402 hasConceptScore W3013559402C154945302 @default.