FPGA-based CNN Processor with Filter-Wise-Optimized Bit Precision.
Asuka MakiDaisuke MiyashitaKengo NakataFumihiko TachibanaTomoya SuzukiJun DeguchiPublished in: A-SSCC (2018)
Keyphrases
- random access memory
- cellular neural networks
- high precision
- convolutional neural network
- high speed
- application specific
- pairwise
- hardware implementation
- parallel processing
- filtering algorithm
- precision and recall
- magnetic tape
- fpga device
- processor core
- multiscale
- median filter
- parallel processors
- parallel architectures
- multiresolution