StyleGAN2-ADA and Real-ESRGAN: Thai font generation with generative adversarial networks

• Published in: Advances in computational intelligence Vol. 4; no. 1; p. 2
• Main Authors: Nitisukanan, Nidchapan, Boonthaweechok, Chotika, Tiawpanichkij, Prapatsorn, Pissakul, Juthamas, Maneesawangwong, Naliya, Siriborvornratanakul, Thitirat
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.03.2024; Springer Nature B.V
• Subjects: Artificial Intelligence; Computational Intelligence; Datasets; Deep learning; Design; Engineering; Federal legislation; Generative adversarial networks; Image enhancement; Image quality; Machine Learning; Original Article; Deep learning; Super-resolution; Font generation; Generative adversarial network
• ISSN: 2730-7794; 2730-7808
• DOI: 10.1007/s43674-024-00069-3

Contemporary font design is a labor-intensive process. To address this, we utilize deep learning, specifically StyleGAN2-ADA and Real-ESRGAN, for automated Thai font generation. StyleGAN2-ADA incorporates adaptive discriminator augmentation (ADA) for image synthesis. By integrating Real-ESRGAN, font quality is enhanced. Our approach produces diverse, high-resolution fonts, as demonstrated in comparative experiments. In a survey with 50 participants, StyleGAN2-ADA without augmentation proves superior in legibility and visual appeal, while StyleGAN2-ADA with augmentation excels in diversity. This research highlights the efficiency of deep learning in creating high-quality Thai fonts and has implications for automated font design advancement.