Batik pattern synthesis for virtual try on application

Chen, Sin Yee (2024) Batik pattern synthesis for virtual try on application. Final Year Project, UTAR.

Preview

PDF Download (3759Kb) | Preview

Abstract

Fabric pattern design is a dynamic blend of culture, tradition, and human ingenuity. It serves as a canvas reflecting cultural identity and aesthetics across various societies, conveying stories, beliefs, and a sense of belonging. However, the labour-intensive and time-consuming nature of traditional pattern design hampers creativity and customization. Generative Adversarial Networks (GANs), a subset of deep learning techniques, offer a promising solution to this challenge. By automating the design process, reducing human intervention, and enhancing the accuracy and realism of generated patterns, GANs bridge the gap between tradition and modernity. However, current GAN-based approaches suffer from inconsistencies, diversity limitations, and control issues. This study aims to leverage GANs to automate fabric pattern generation while preserving cultural heritage. It seeks to enhance GAN architectures, diversify training datasets, and gradually increase pattern complexity. This research empowers designers to create unique fabric patterns, revitalizes traditional designs, and enriches the fashion industry with culturally rich fabric models. Ultimately, it paves the way for a new era of efficient, creative, and culturally significant fabric pattern design.

Actions (login required)