Yejin Jin, Janghwan Kim, Hyunseung Son, R. Young Chul Kim, Journal of Internet Computing and Services, Vol. 27, No. 1, pp. 39-47, Feb. 2026
10.7472/jksii.2026.27.1.39, Full Text:  HTML
Keywords: GPT, Natural Language Processing, State Diagram, 3D modeling, Cartoon Engineering

Abstract

With rapid growth in the 3D Content market, there is an increasing demand for 3D technology to produce 3D content. Recently, in the cartoon industry, cartoonists are attempting to utilize AI image tools to generate background elements or props for cartoons using natural language sentences. However, Generative AI faces technical difficulties in maintaining the visual consistency of main characters across all scenes. Therefore, a methodology is required that ensures the consistent reuse of main characters throughout the cartoon story. To address this, we propose an automated mechanism for representing a 3D object's emotional expression by analyzing natural language-based cartoon sentences. This enables the emotional state of the main character in the cartoon to be reflected in the 3D character entity. The existing Text-to-3D tools generated various models for each natural language query, which cannot guarantee the accuracy of the results. To overcome these limitations, natural language sentences are analyzed using linguistic approaches via the GPT API. Subsequently, cartoon sentences are visualized through UML diagrams, integrating software engineering principles into cartoon engineering. By ensuring reusability and consistency for 3D characters, the proposed method is expected to enhance the productivity and convenience of 3D cartoon development.

Statistics
Cite this article