High-Performance FPGA-Based CNN Accelerator With Block-Floating-Point Arithmetic.
Xiaocong LianZhenyu LiuZhourui SongJiwu DaiWei ZhouXiangyang JiPublished in: IEEE Trans. Very Large Scale Integr. Syst. (2019)
Keyphrases
- floating point arithmetic
- floating point
- field programmable gate array
- compute intensive
- cellular neural networks
- hardware implementation
- application specific
- database
- convolutional neural network
- general purpose
- fixed point
- instruction set
- parallel implementation
- image processing algorithms
- image blocks
- parallel computing
- embedded systems
- parallel processing
- image segmentation
- information systems