Convolutional neural network (CNN) has achieved impressive success in computer vision during the past few decades. The image convolution operation helps CNNs to get good performance on image-related tasks. However, the image convolution has high computation complexity and hard to be implemented. This paper proposes the CEMNet, which can be trained in the frequency domain. The most important motivation of this research is that we can use the straightforward element-wise multiplication operation to replace the image convolution in the frequency domain based on the Cross-Correlation Theorem, which obviously reduces the computation complexity. We further introduce a Weight Fixation mechanism to alleviate the problem of over-fitting, and analyze the working behavior of Batch Normalization, Leaky ReLU, and Dropout in the frequency domain to design their counterparts for CEMNet. Also, to deal with complex inputs brought by Discrete Fourier Transform, we design a two-branches network structure for CEMNet. Experimental results imply that CEMNet achieves good performance on MNIST and CIFAR-10 databases.
翻译:在过去几十年中,图像变异操作在计算机视觉方面取得了令人印象深刻的成功。图像变异操作帮助CNN在图像相关任务上取得了良好的表现。然而,图像变异具有很高的计算复杂性,难以执行。本文件提议CEMNet,可以在频率域内进行培训。这一研究的最重要动机是,我们可以使用直接的元素-介质倍增操作来取代基于交叉校正理论的频率域内图像变异,这显然降低了计算的复杂性。我们进一步引入了一种重力修正机制,以减轻超装问题,并分析Batch Alcardization、Leky ReLU和在频率域内脱机的工作行为,以设计CEMNet的对应方。此外,为了处理Discrete Fourier变换带来的复杂投入,我们为CEMNet设计了一个两层网络结构。实验结果表明,CEMNet在MNIST和CIFAR-10数据库上取得了良好的业绩。