Najdi Threads

Inspiration Our project, Najdi Threads, was inspired by the rich cultural heritage of the Najd region in Saudi Arabia. We were particularly fascinated by the intricate embroidery, unique color palettes, and symbolism found in traditional garments worn by men and women in the region. We wanted to preserve this artistic heritage by blending it with modern AI tools. What We Learned Through this project, we learned how to train and fine-tune image generation models to produce art rooted in cultural aesthetics. We also explored how AI can be used as a tool to preserve identity, heritage, and storytelling. How We Built It We built the project using a combination of Python and PyTorch, utilizing a GAN (Generative Adversarial Network) architecture. We trained the model on a dataset of Najdi clothing images, including thobes, abayas, and accessories. The generator creates new visuals inspired by these inputs, while the discriminator ensures quality and authenticity in design. Challenges We Faced One major challenge was the limited availability of labeled datasets for traditional clothing. We had to collect and curate images ourselves. Another challenge was training stability, which required careful tuning of parameters and batch sizes. Despite this, we achieved a satisfying result that represents the spirit of Najdi tradition in a modern digital form.

Built With

• google-colab
• markdown
• matplotlib
• numpy
• pillow
• python
• pytorch