Abstract:
This work investigates the use of Neural Style Transfer (NST) with the VGG19 network
to achieve artistic manipulation of natural images. NST offers a powerful technique to
create new images by combining the content of a photograph (content image) with the
style of another image (style image). This research leverages the VGG19 model’s ability to
differentiate between content and style features within images. By minimizing the content
and style distances between the generated image, the content image, and the style image, the
proposed method allows for transforming various image types. This opens the possibility
of applying artistic styles like oil paintings to user-provided images or even live camera
captures. Furthermore, the work explores the development of a user-friendly interface for
this NST application, potentially promoting creative image manipulation within the images.
