FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2906403229> ?p ?o ?g. }

• W2906403229 endingPage "22" @default.
• W2906403229 startingPage "1" @default.
• W2906403229 abstract "Convolutional Neural Networks-- (CNNs) based algorithms have been successful in solving image recognition problems, showing very large accuracy improvement. In recent years, deconvolution layers are widely used as key components in the state-of-the-art CNNs for end-to-end training and models to support tasks such as image segmentation and super resolution. However, the deconvolution algorithms are computationally intensive, which limits their applicability to real-time applications. Particularly, there has been little research on the efficient implementations of deconvolution algorithms on FPGA platforms that have been widely used to accelerate CNN algorithms by practitioners and researchers due to their high performance and power efficiency. In this work, we propose and develop deconvolution architecture for efficient FPGA implementation. FPGA-based accelerators are proposed for both deconvolution and CNN algorithms. Besides, memory sharing between the computation modules is proposed for the FPGA-based CNN accelerator as well as for other optimization techniques. A non-linear optimization model based on the performance model is introduced to efficiently explore the design space to achieve optimal processing speed of the system and improve power efficiency. Furthermore, a hardware mapping framework is developed to automatically generate the low-latency hardware design for any given CNN model on the target device. Finally, we implement our designs on Xilinx Zynq ZC706 board and the deconvolution accelerator achieves a performance of 90.1 giga operations per second (GOPS) under 200MHz working frequency and a performance density of 0.10 GOPS/DSP using 32-bit quantization, which significantly outperforms previous designs on FPGAs. A real-time application of scene segmentation on Cityscapes Dataset is used to evaluate our CNN accelerator on Zynq ZC706 board, and the system achieves a performance of 107 GOPS and 0.12 GOPS/DSP using 16-bit quantization and supports up to 17 frames per second for 512 × 512 image inputs with a power consumption of only 9.6W." @default.
• W2906403229 created "2019-01-01" @default.
• W2906403229 creator A5054407201 @default.
• W2906403229 creator A5056354430 @default.
• W2906403229 creator A5056495185 @default.
• W2906403229 creator A5057043409 @default.
• W2906403229 creator A5057940557 @default.
• W2906403229 creator A5061397387 @default.
• W2906403229 date "2018-09-30" @default.
• W2906403229 modified "2023-10-03" @default.
• W2906403229 title "Optimizing CNN-based Segmentation with Deeply Customized Convolutional and Deconvolutional Architectures on FPGA" @default.
• W2906403229 cites W1745334888 @default.
• W2906403229 cites W1903029394 @default.
• W2906403229 cites W1932847118 @default.
• W2906403229 cites W2094756095 @default.
• W2906403229 cites W2141200610 @default.
• W2906403229 cites W2155893237 @default.
• W2906403229 cites W2164700406 @default.
• W2906403229 cites W2276486856 @default.
• W2906403229 cites W2293078015 @default.
• W2906403229 cites W2340897893 @default.
• W2906403229 cites W2476548250 @default.
• W2906403229 cites W2508457857 @default.
• W2906403229 cites W2553040026 @default.
• W2906403229 cites W2585720638 @default.
• W2906403229 cites W2597692291 @default.
• W2906403229 cites W2625457103 @default.
• W2906403229 cites W2729070998 @default.
• W2906403229 cites W2729080111 @default.
• W2906403229 cites W2741543856 @default.
• W2906403229 cites W2891976760 @default.
• W2906403229 cites W2963363373 @default.
• W2906403229 cites W2963577671 @default.
• W2906403229 cites W2963881378 @default.
• W2906403229 doi "https://doi.org/10.1145/3242900" @default.
• W2906403229 hasPublicationYear "2018" @default.
• W2906403229 type Work @default.
• W2906403229 sameAs 2906403229 @default.
• W2906403229 citedByCount "40" @default.
• W2906403229 countsByYear W29064032292018 @default.
• W2906403229 countsByYear W29064032292019 @default.
• W2906403229 countsByYear W29064032292020 @default.
• W2906403229 countsByYear W29064032292021 @default.
• W2906403229 countsByYear W29064032292022 @default.
• W2906403229 countsByYear W29064032292023 @default.
• W2906403229 crossrefType "journal-article" @default.
• W2906403229 hasAuthorship W2906403229A5054407201 @default.
• W2906403229 hasAuthorship W2906403229A5056354430 @default.
• W2906403229 hasAuthorship W2906403229A5056495185 @default.
• W2906403229 hasAuthorship W2906403229A5057043409 @default.
• W2906403229 hasAuthorship W2906403229A5057940557 @default.
• W2906403229 hasAuthorship W2906403229A5061397387 @default.
• W2906403229 hasBestOaLocation W29064032292 @default.
• W2906403229 hasConcept C113775141 @default.
• W2906403229 hasConcept C11413529 @default.
• W2906403229 hasConcept C114614502 @default.
• W2906403229 hasConcept C149635348 @default.
• W2906403229 hasConcept C154945302 @default.
• W2906403229 hasConcept C173608175 @default.
• W2906403229 hasConcept C174576160 @default.
• W2906403229 hasConcept C2776221188 @default.
• W2906403229 hasConcept C28855332 @default.
• W2906403229 hasConcept C33923547 @default.
• W2906403229 hasConcept C41008148 @default.
• W2906403229 hasConcept C42935608 @default.
• W2906403229 hasConcept C45347329 @default.
• W2906403229 hasConcept C50644808 @default.
• W2906403229 hasConcept C74193536 @default.
• W2906403229 hasConcept C81363708 @default.
• W2906403229 hasConcept C84462506 @default.
• W2906403229 hasConcept C9390403 @default.
• W2906403229 hasConceptScore W2906403229C113775141 @default.
• W2906403229 hasConceptScore W2906403229C11413529 @default.
• W2906403229 hasConceptScore W2906403229C114614502 @default.
• W2906403229 hasConceptScore W2906403229C149635348 @default.
• W2906403229 hasConceptScore W2906403229C154945302 @default.
• W2906403229 hasConceptScore W2906403229C173608175 @default.
• W2906403229 hasConceptScore W2906403229C174576160 @default.
• W2906403229 hasConceptScore W2906403229C2776221188 @default.
• W2906403229 hasConceptScore W2906403229C28855332 @default.
• W2906403229 hasConceptScore W2906403229C33923547 @default.
• W2906403229 hasConceptScore W2906403229C41008148 @default.
• W2906403229 hasConceptScore W2906403229C42935608 @default.
• W2906403229 hasConceptScore W2906403229C45347329 @default.
• W2906403229 hasConceptScore W2906403229C50644808 @default.
• W2906403229 hasConceptScore W2906403229C74193536 @default.
• W2906403229 hasConceptScore W2906403229C81363708 @default.
• W2906403229 hasConceptScore W2906403229C84462506 @default.
• W2906403229 hasConceptScore W2906403229C9390403 @default.
• W2906403229 hasIssue "3" @default.
• W2906403229 hasLocation W29064032291 @default.
• W2906403229 hasLocation W29064032292 @default.
• W2906403229 hasOpenAccess W2906403229 @default.
• W2906403229 hasPrimaryLocation W29064032291 @default.
• W2906403229 hasRelatedWork W1535188164 @default.
• W2906403229 hasRelatedWork W2349666940 @default.
• W2906403229 hasRelatedWork W2352017551 @default.
• W2906403229 hasRelatedWork W2375677847 @default.

• next