Matches in SemOpenAlex for { <https://semopenalex.org/work/W2916321174> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W2916321174 abstract "Convolutional neural networks (CNN) have demonstrated state-of-the-art accuracy in image classification and object detection owing to the increase in data and computation capacity of hardware. However, this state-of-the-art achievement depends heavily on the DSP floating-point computing capability of the device, which increases the power dissipation and cost of the device. In order to solve the problem, we made the first attempt to implement a CNN computing accelerator based on shift operation on FPGA. In this accelerator, an efficient Incremental Network Quantization (INQ) method was applied to compress the CNN model from full precision to 4-bit integer, which represents values of either zero or power of two. Then the multiply and accumulate (MAC) operations for convolution layer and fully-connected layer was converted to shift and accumulation (SAC) operations, and SAC could be easily implemented by the logic elements of FPGA. Consequently, parallelism of CNN inference process can be further expanded. For the SqueezeNet model, single image processing latency was 0.673ms on Intel Arria 10 FPGA (Inspur F10A board) showing a slightly better result than on NVIDIA Tesla P4, and the compute capacity of FPGA increased by 1.77 times at least." @default.
- W2916321174 created "2019-03-02" @default.
- W2916321174 creator A5027835055 @default.
- W2916321174 creator A5037904266 @default.
- W2916321174 creator A5085836223 @default.
- W2916321174 date "2019-02-20" @default.
- W2916321174 modified "2023-09-23" @default.
- W2916321174 title "A Deep Learning Inference Accelerator Based on Model Compression on FPGA" @default.
- W2916321174 cites W2094756095 @default.
- W2916321174 cites W2109835049 @default.
- W2916321174 cites W2153039279 @default.
- W2916321174 cites W2318533486 @default.
- W2916321174 cites W2395355464 @default.
- W2916321174 cites W2527036487 @default.
- W2916321174 cites W2574797063 @default.
- W2916321174 cites W2950656546 @default.
- W2916321174 cites W2962676243 @default.
- W2916321174 doi "https://doi.org/10.1145/3289602.3293938" @default.
- W2916321174 hasPublicationYear "2019" @default.
- W2916321174 type Work @default.
- W2916321174 sameAs 2916321174 @default.
- W2916321174 citedByCount "2" @default.
- W2916321174 countsByYear W29163211742019 @default.
- W2916321174 crossrefType "proceedings-article" @default.
- W2916321174 hasAuthorship W2916321174A5027835055 @default.
- W2916321174 hasAuthorship W2916321174A5037904266 @default.
- W2916321174 hasAuthorship W2916321174A5085836223 @default.
- W2916321174 hasConcept C108583219 @default.
- W2916321174 hasConcept C11413529 @default.
- W2916321174 hasConcept C114614502 @default.
- W2916321174 hasConcept C13164978 @default.
- W2916321174 hasConcept C154945302 @default.
- W2916321174 hasConcept C173608175 @default.
- W2916321174 hasConcept C2776214188 @default.
- W2916321174 hasConcept C28855332 @default.
- W2916321174 hasConcept C33923547 @default.
- W2916321174 hasConcept C41008148 @default.
- W2916321174 hasConcept C42935608 @default.
- W2916321174 hasConcept C45347329 @default.
- W2916321174 hasConcept C45374587 @default.
- W2916321174 hasConcept C50644808 @default.
- W2916321174 hasConcept C74193536 @default.
- W2916321174 hasConcept C76155785 @default.
- W2916321174 hasConcept C81363708 @default.
- W2916321174 hasConcept C82876162 @default.
- W2916321174 hasConcept C84211073 @default.
- W2916321174 hasConcept C84462506 @default.
- W2916321174 hasConcept C9390403 @default.
- W2916321174 hasConceptScore W2916321174C108583219 @default.
- W2916321174 hasConceptScore W2916321174C11413529 @default.
- W2916321174 hasConceptScore W2916321174C114614502 @default.
- W2916321174 hasConceptScore W2916321174C13164978 @default.
- W2916321174 hasConceptScore W2916321174C154945302 @default.
- W2916321174 hasConceptScore W2916321174C173608175 @default.
- W2916321174 hasConceptScore W2916321174C2776214188 @default.
- W2916321174 hasConceptScore W2916321174C28855332 @default.
- W2916321174 hasConceptScore W2916321174C33923547 @default.
- W2916321174 hasConceptScore W2916321174C41008148 @default.
- W2916321174 hasConceptScore W2916321174C42935608 @default.
- W2916321174 hasConceptScore W2916321174C45347329 @default.
- W2916321174 hasConceptScore W2916321174C45374587 @default.
- W2916321174 hasConceptScore W2916321174C50644808 @default.
- W2916321174 hasConceptScore W2916321174C74193536 @default.
- W2916321174 hasConceptScore W2916321174C76155785 @default.
- W2916321174 hasConceptScore W2916321174C81363708 @default.
- W2916321174 hasConceptScore W2916321174C82876162 @default.
- W2916321174 hasConceptScore W2916321174C84211073 @default.
- W2916321174 hasConceptScore W2916321174C84462506 @default.
- W2916321174 hasConceptScore W2916321174C9390403 @default.
- W2916321174 hasLocation W29163211741 @default.
- W2916321174 hasOpenAccess W2916321174 @default.
- W2916321174 hasPrimaryLocation W29163211741 @default.
- W2916321174 hasRelatedWork W2915614342 @default.
- W2916321174 hasRelatedWork W2949649078 @default.
- W2916321174 hasRelatedWork W3005273712 @default.
- W2916321174 hasRelatedWork W3014927913 @default.
- W2916321174 hasRelatedWork W3015861875 @default.
- W2916321174 hasRelatedWork W4214908790 @default.
- W2916321174 hasRelatedWork W4312417841 @default.
- W2916321174 hasRelatedWork W4317495558 @default.
- W2916321174 hasRelatedWork W4322776108 @default.
- W2916321174 hasRelatedWork W4361251788 @default.
- W2916321174 isParatext "false" @default.
- W2916321174 isRetracted "false" @default.
- W2916321174 magId "2916321174" @default.
- W2916321174 workType "article" @default.