Show simple item record

dc.contributor.authorRecoskie, Daniel
dc.contributor.authorMann, Richard 19:08:32 (GMT) 19:08:32 (GMT)
dc.description©2018 IEEEen
dc.description.abstractWe propose a new method for learning filters for the 2D discrete wavelet transform. We extend our previous work on the 1D wavelet transform in order to process images. We show that the 2D wavelet transform can be represented as a modified convolutional neural network (CNN). Doing so allows us to learn wavelet filters from data by gradient descent. Our learned wavelets are similar to traditional wavelets which are typically derived using Fourier methods. For filter comparison, we make use of a cosine measure under all filter rotations. The learned wavelets are able to capture the structure of the training data. Furthermore, we can generate images from our model in order to evaluate the filters. The main findings of this work is that wavelet functions can arise naturally from data, without the need for Fourier methods. Our model requires relatively few parameters compared to traditional CNNs, and is easily incorporated into neural network frameworks.en
dc.description.sponsorshipNatural Sciences and Engineering Research Council of Canadaen
dc.subjectconvolution neural networken
dc.subjectfilter banksen
dc.titleLearning Filters for the 2D Wavelet Transformen
dc.typeConference Paperen
dcterms.bibliographicCitationRecoskie, Daniel, and Richard Mann. ‘Learning Filters for the 2D Wavelet Transform’. In 2018 15th Conference on Computer and Robot Vision (CRV), 198–205, 2018.
uws.contributor.affiliation1Faculty of Mathematicsen
uws.contributor.affiliation2David R. Cheriton School of Computer Scienceen

Files in this item


This item appears in the following Collection(s)

Show simple item record


University of Waterloo Library
200 University Avenue West
Waterloo, Ontario, Canada N2L 3G1
519 888 4883

All items in UWSpace are protected by copyright, with all rights reserved.

DSpace software

Service outages