CNN-DMA: A Predictable and Scalable Direct Memory Access Engine for Convolutional Neural Network with Sliding-window Filtering.
Zheng WangZhuo WangJian LiaoChao ChenYongkui YangBo DongWeiguang ChenWenxuan ChenMing LeiWeiyu GuoRui ChenYi PengZhibin YuPublished in: ACM Great Lakes Symposium on VLSI (2021)
Keyphrases
- convolutional neural network
- sliding window
- direct memory access
- data streams
- face detection
- limited memory
- operating system
- fixed size
- tcp ip
- window size
- main memory
- high speed
- stream data
- streaming data
- variable size
- neural network
- continuous queries
- object detection
- general purpose
- real time
- information systems
- data acquisition
- computer vision
- index structure
- face recognition