How Generative AI is Reshaping Traditional Japanese Art

71
1
How Generative AI is Reshaping Traditional Japanese Art

The Evo-Ukiyoe system, a brainchild of Tarin Clanuwat, stands out as a groundbreaking innovation in the realm of traditional Japanese art. Unlike conventional AI models that often fail to capture the distinct style of ukiyo-e, Evo-Ukiyoe uses a vast database of over 24,000 scanned pieces to churn out woodblock prints based on user prompts, bringing a modern twist to the historical genre. Clanuwat's vision for this AI model is to introduce classics to the general public in their daily lives, fostering a renewed appreciation for this ancient art form.

In a collaborative effort with Ryo Akama, a literature professor at Ritsumeikan University, Clanuwat aims to create an AI system that perfectly replicates ukiyo-e's authentic color variations and textures. Utilizing her background in computer programming and experience in Google's AI research and development, Clanuwat embarked on a mission to bridge the gap between classical art and modern technology. This endeavor led to the development of Evo-Ukiyoe, allowing users to witness the fusion of artistry and artificial intelligence in real-time creations.

The AI art generator, Evo-Ukiyoe, not only accurately captures the essence of traditional ukiyo-e prints but also encapsulates the genre's ambiance through its depiction of humans. By incorporating elements such as limited hues and woodblock print characteristics, the system seamlessly blends past and present aesthetics. While some challenges persist, like occasional depictions of men in women's kimono, Clanuwat remains dedicated to refining the AI model to enhance its understanding of complex art forms such as kimono design and hairstyles, demonstrating her commitment to artistic integrity and historical accuracy.