1. 论文思想
一维滤过器。将三维卷积分解成三个一维卷积。convolution across channels(lateral), vertical and horizontal direction.一维滤过器。将三维卷积分解成三个一维卷积。convolution acro
阅读背景：
论文笔记—Flattened convolution neural networks for feedforward acceleration
来源：互联网
分享到：
