Wavelet pooling for convolutional neural networks

Abstract

[en] Wavelets are mathematical functions that are currently used in many computer vision problems, such as image denoising or image compression. In this work, first we will study all the basic theory about wavelets, in order to understand them and build a basic knowledge that allows us to develop another application. For such purpose, we propose two pooling methods based on wavelets: one based on simple wavelet basis and one that combines two basis working in parallel. We will test them and show that they can be used at the same level of performance as max and average pooling.