This AI Paper Proposes Two Types of Convolution, Pixel Difference Convolution (PDC) and Binary Pixel Difference Convolution (Bi-PDC), to Enhance the Representation Capacity of Convolutional Neural Network CNNs

This AI Paper Proposes Two Types of Convolution, Pixel Difference Convolution (PDC) and Binary Pixel Difference Convolution (Bi-PDC), to Enhance the Representation Capacity of Convolutional Neural Network CNNs

by | Feb 13, 2024 | Uncategorized

Deep convolutional neural networks (DCNNs) have been a game-changer for several computer vision tasks. These include object identification, object recognition, image segmentation, and edge detection. The ever-growing size and power consumption of DNNs have been key to...