Evolution of Winning Solutions in the 2021 Low-Power Computer Vision Challenge

Xiao Hu, Ziteng Jiao, Ayden Kocher, Zhenyu Wu, Junjie Liu, James C. Davis, George K. Thiruvathukal, Yung-Hsiang Lu

Abstract: We organized the 2021 IEEE Low-Power Computer Vision Challenge to advance state-of-the-art solutions in low-power computer vision. Here, we examine the winning teams’ development patterns and design decisions, focusing on their techniques to balance power consumption and accuracy to provide guidelines for future competitions.