GaborNet: Gabor filters with learnable parameters in deep convolutional neural network

作者：Andrey Sergeevich Alekseev, Anatoly Bobe · 发表于：2019 International Conference on Engineering and Telecommunication (EnT) · 年份：2019 · DOI：10.1109/EnT47717.2019.9030571 · 被引用次数：91 · 研究领域：Computer Science、Engineering

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.