In this paper, we propose a novel network design mechanism for efficient embedded computing. Inspired by the limited computing patterns, we propose to fix the number of channels in a group convolution, instead of the existing practice that fixing the total group numbers. Our solution based network, named Variable Group Convolutional Network (VarGNet), can be optimized easier on hardware side, due to the more unified computing schemes among the layers. Extensive experiments on various vision tasks, including classification, detection, pixel-wise parsing and face recognition, have demonstrated the practical value of our VarGNet.
@article{arxiv.1907.05653,
title = {VarGNet: Variable Group Convolutional Neural Network for Efficient Embedded Computing},
author = {Qian Zhang and Jianjun Li and Meng Yao and Liangchen Song and Helong Zhou and Zhichao Li and Wenming Meng and Xuezhi Zhang and Guoli Wang},
journal= {arXiv preprint arXiv:1907.05653},
year = {2020}
}