A CNN Accelerator on FPGA Using Depthwise Separable Convolution.
Lin BaiYiming ZhaoXinming HuangPublished in: CoRR (2018)
Keyphrases
- field programmable gate array
- cellular neural networks
- hardware implementation
- parallel computing
- image processing algorithms
- embedded systems
- image processing
- convolutional neural network
- hardware design
- fpga implementation
- hardware architecture
- computing systems
- low power consumption
- convolution kernel
- fpga technology
- software implementation
- centroid neural network
- high speed
- massively parallel
- single chip
- parallel implementation
- fpga device
- digital signal
- dedicated hardware
- verilog hdl
- hardware software
- real time image processing
- linearly separable
- fast fourier transform
- face detection
- fir filters
- hardware architectures
- signal processing
- neural network
- real time