GaborNet: Gabor filters with learnable parameters in deep convolutional neural networks
Computer Vision and Pattern Recognition
2019-05-01 v1 Machine Learning
Image and Video Processing
Abstract
The article describes a system for image recognition using deep convolutional neural networks. Modified network architecture is proposed that focuses on improving convergence and reducing training complexity. The filters in the first layer of the network are constrained to fit the Gabor function. The parameters of Gabor functions are learnable and are updated by standard backpropagation techniques. The system was implemented on Python, tested on several datasets and outperformed the common convolutional networks.
Cite
@article{arxiv.1904.13204,
title = {GaborNet: Gabor filters with learnable parameters in deep convolutional neural networks},
author = {Andrey Alekseev and Anatoly Bobe},
journal= {arXiv preprint arXiv:1904.13204},
year = {2019}
}
Comments
10 pages, 6 figures, 3 tables, preprint