IP Core for Efficient Zero-Run Length Compression of CNN Feature Maps
Convolutional Neural Networks (CNNs) are becoming a fundamental tool for machine learning. High performance and energy efficiency are of great importance for deployments of CNNs in many embedded applications. Energy consumption during CNN processing is dominated by memory access and since large netw...
Main Authors: | A. Erdeljan, B. Vukobratović, R. Struharik |
---|---|
Format: | Article |
Language: | English |
Published: |
Telecommunications Society, Academic Mind
2018-06-01
|
Series: | Telfor Journal |
Subjects: | |
Online Access: |
http://journal.telfor.rs/Published/Vol10No1/Vol10No1_A8.pdf
|
Similar Items
-
HW/SW Architecture Exploration for an Efficient Implementation of the Secure Hash Algorithm SHA-256
by: Manel Kammoun, et al.
Published: (2021-06-01) -
Design and implementation of the BSP for automated monitoring system based on Zynq-7000
by: Song Kai, et al.
Published: (2018-09-01) -
System verilog RTL modeling with embedded assertions[electronic resource] /
by: Chow, Chee Siang, 1984-, et al.
Published: (2012) -
System verilog RTL modeling with embedded assertions /
by: Chow, Chee Siang, 1984-, et al.
Published: (2012) -
Development of neutron flux measurement module for research reactors based on FPGA. Part-1-Model development and generic verification
by: Mjzoob I. Ibrahim, et al.
Published: (2020-12-01)