Description
Energy efficiency is critical for running computer vision on battery-powered systems, such as mobile phones or UAVs (unmanned aerial vehicles, or drones). This book collects the methods that have won the annual IEEE Low-Power Computer Vision Challenges since 2015. The winners share their solutions and provide insight on how to improve the efficiency of machine learning systems.
Table of Contents
Section I Introduction
Book Introduction
Yung-Hsiang Lu, George K. Thiruvathukal, Jaeyoun Kim, Yiran Chen, and Bo Chen
History of Low-Power Computer Vision Challenge
Yung-Hsiang Lu and Xiao Hu, Yiran Chen, Joe Spisak, Gaurav Aggarwal, Mike Zheng Shou, and George K. Thiruvathukal
Survey on Energy-Efficient Deep Neural Networks for Computer Vision
Abhinav Goel, Caleb Tung, Xiao Hu, Haobo Wang, and Yung-Hsiang Lu and George K. Thiruvathukal
Section II Competition Winners
Hardware design and software practices for efficient neural network inference
Yu Wang, Xuefei Ning, Shulin Zeng, Yi Kai, Kaiyuan Guo, and Hanbo Sun, Changcheng Tang, Tianyi Lu, Shuang Liang, and Tianchen Zhao
Progressive Automatic Design of Search Space for One-Shot Neural Architecture Search
Xin Xia, Xuefeng Xiao, and Xing Wang
Fast Adjustable Threshold For Uniform Neural Network Quantization
Alexander Goncharenko, Andrey Denisov, and Sergey Alyamkin
Power-efficient Neural Network Scheduling on Heterogeneous SoCs
Ying Wang, Xuyi Cai, and Xiandong Zhao
Efficient Neural Network Architectures
Han Cai and Song Han
Design Methodology for Low Power Image Recognition Systems
Soonhoi Ha, EunJin Jeong, Duseok Kang, Jangryul Kim, and Donghyun Kang
Guided Design for Efficient On-device Object Detection Model
Tao Sheng and Yang Liu
Section III Invited Articles
Quantizing Neural Networks
Marios Fournarakis, Markus Nagel, Rana Ali Amjad, Yelysei Bondarenko, Mart van Baalen, and Tijmen Blankevoort
A practical guide to designing efficient mobile architectures
Mark Sandler and Andrew Howard
A Survey of Quantization Methods for Efficient Neural Network Inference
Amir Gholami, Sehoon Kim, Zhen Dong, Zhewei Yao, Michael Mahoney, and Kurt Keutzer
Bibliography
Index
