Artificial Intelligence-Driven Character Design in Digital Graphics through Technical Solutions and Innovations
摘要
The study examines the potential of neural networks as catalysts for creative work and as expanders of the horizons of possibility in the art of comic book creation. In this work, the authors analyze current approaches and perspectives on integrating intelligent machine methods into character construction for graphic narratives. Particular attention is given to the role of neurotechnologies in optimizing the production of a visual character, including the original concept of character creation called “Oxyumoron,” which reflects the synthesis of opposites (“The Binomial of Imagination”), a fundamental element of the proposed algorithmic cycle of generation and integration of character structures into the artistic fabric of the comic. The authors conducted experiments to implement these ideas using advanced generative creativity models, such as Generative Adversarial Networks, Convolutional Neural Network architectures, and transformers, to generate fundamentally new character images. Based on the study’s results, the authors emphasize the importance of a conscious, balanced approach to integrating artificial intelligence into the creation of expressive, emotionally rich characters that enrich the artistic world of comics and engage readers with the image’s depth.