A Low-Power Sparse Convolutional Neural Network Accelerator With Pre-Encoding Radix-4 Booth Multiplier.
Quan ChengLiuyao DaiMingqiang HuangAo ShenWei MaoMasanori HashimotoHao YuPublished in: IEEE Trans. Circuits Syst. II Express Briefs (2023)
Keyphrases
- low power
- convolutional neural network
- floating point
- low cost
- power consumption
- high speed
- face detection
- high power
- vlsi circuits
- single chip
- vlsi architecture
- wireless transmission
- digital signal processing
- logic circuits
- low power consumption
- mixed signal
- neural network
- low density parity check
- parallel implementation
- hardware implementation
- multi class
- computer vision
- cmos technology
- image sensor
- delay insensitive
- pairwise
- real time