Title: Neural Style Transfer Using Convolutional Neural Networks: An Experimental Study
Authors: Alaa Yousef Abu Sultan, Samy S. Abu-Naser
Volume: 10
Issue: 2
Pages: 50-57
Publication Date: 2026/02/28
Abstract:
This paper presents an experimental study on neural style transfer, a deep learning technique that synthesizes a new image by combining the semantic content of one image with the artistic style of another. A pretrained convolutional neural network (CNN) is employed to transform a natural landscape photograph into a stylized image characterized by expressive brushstrokes and enhanced texture patterns. The experiment demonstrates how deep neural networks can disentangle and recombine content and style representations extracted from different layers of the network. The resulting output preserves the essential structural features of the original scene while adopting stylistic properties reminiscent of post-impressionist artwork. The findings highlight the effectiveness of pretrained CNNs for artistic image synthesis and creative visual transformation.