NEURAL STYLE TRANSFER AS DIGITAL REPRESENTATION THROUGH PAINTING USING DEEP LEARNING

The intersection of art and technology, particularly the development of deep learning and artificial intelligence, has opened new avenues for creative expression. Neural Style Transfer (NST) is a new approach that uses deep neural networks to add creative style to digital images. This technique overlays visual elements—such as texture, color patterns, and brushstrokes—from one image onto the content of another, producing interesting and imaginative results. This study explores the fundamental concepts, methodology, and artistic significance of neural style transfer. This technique is primarily based on convolutional neural networks (CNNs), which are capable of extracting high-level content features and low-level style features from images. Through an optimization process that separates and then reassembles these features, NST creates images that retain the structural content of the original image while adopting the artistic style of the reference artwork. This research highlights the interdisciplinary impact of neural style transfer, demonstrating its relevance not only in digital art and graphic design, but also in fields such as animation, multimedia, and visual communication. Furthermore, NST serves as a bridge between traditional artistic methods and modern computational techniques, allowing artists, designers, and technologists to collaborate and explore new creative possibilities. The results show that neural style transfer is an important step toward integrating deep learning.