What is it about?

We are exploring calligraphic Arabic script, particularly the "nastaliq" style common in Iran, using Generative Adversarial Networks (GANs). Given the challenges of the cursive nature of Arabic script and the lack of conventional tools to represent it, our work aims to blend traditional calligraphy with modern technology. We have created two datasets, Nas4-60k and Nas4-60k-aug, to train our generative models in producing calligraphy. Using the StyleGAN2-ada architecture, our system successfully generates high-quality calligraphic samples that extend beyond the training data. These samples demonstrate a wide range of meaningful features, blending traditional and contemporary elements of calligraphic Arabic. Inspired by the compositional form of "siyah-mashq," our project has resulted in several publicly presented artworks, showcasing a new way of creative expression through the use of GANs in calligraphic Arabic script.

Featured Image

Read the Original

This page is a summary of: Unveiling New Artistic Dimensions in Calligraphic Arabic Script with Generative Adversarial Networks, Proceedings of the ACM on Computer Graphics and Interactive Techniques, July 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3664210.
You can read the full text:

Read

Contributors

The following have contributed to this page