Innovative 3D Character Model Texture Mapping Solution Based on Artificial Intelligence Image Generation Model
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This paper presents an innovative solution for 3D character model texture mapping based on artificial intelligence (AI) image generation models. This study explored the potential of AI-generated textures to enhance the efficiency and effectiveness of the traditional manual workflow. By leveraging state-of-the-art AI image generation techniques, we proposed a new workflow that could address existing challenges in the current texture mapping process. This research began with an introduction to AI image generation models with a review of the current status and applications of these models in texture map generation. We then identified and analyzed issues associated with the existing character model texture workflow, highlighting inefficiencies and areas for improvement. Our proposed AI-assisted workflow aimed to streamline the texture creation process, significantly reducing the time required while maintaining high-quality outputs. To validate the new workflow, we conducted a series of experiments with professional 3D artists, yielding clear and significant results. The new AI-assisted workflow significantly reduced texture creation time by an average of 76% to 89% compared to traditional manual methods, demonstrating its potential to drastically improve efficiency. Qualitative feedback from artists highlighted that the new workflow not only streamlined the process, but also made it more user-friendly and accessible. However, some limitations were identified, including the need for further refinement of AI-generated textures to achieve a quality level comparable to those created by highly skilled artists.
Abstract
This paper presents an innovative solution for 3D character model texture mapping based on artificial intelligence (AI) image generation models. This study explored the potential of AI-generated textures to enhance the efficiency and effectiveness of the traditional manual workflow. By leveraging state-of-the-art AI image generation techniques, we proposed a new workflow that could address existing challenges in the current texture mapping process. This research began with an introduction to AI image generation models with a review of the current status and applications of these models in texture map generation. We then identified and analyzed issues associated with the existing character model texture workflow, highlighting inefficiencies and areas for improvement. Our proposed AI-assisted workflow aimed to streamline the texture creation process, significantly reducing the time required while maintaining high-quality outputs. To validate the new workflow, we conducted a series of experiments with professional 3D artists, yielding clear and significant results. The new AI-assisted workflow significantly reduced texture creation time by an average of 76% to 89% compared to traditional manual methods, demonstrating its potential to drastically improve efficiency. Qualitative feedback from artists highlighted that the new workflow not only streamlined the process, but also made it more user-friendly and accessible. However, some limitations were identified, including the need for further refinement of AI-generated textures to achieve a quality level comparable to those created by highly skilled artists.
참고문헌 (11)
- S. Tan, "Animation image art design mode using 3d modeling technology", Wireless Communications and Mobile Computing, 2022, pp.1~9 (2022)[DOI]
- X. Wang, "Evolution and innovations in animation : a comprehensive review and future directions", Concurrency and Computation : Practice and Experience, 36(2) (2023)[DOI]
- X. Liu, "The path of film and television animation creation using virtual reality technology under the artificial intelligence", Scientific Programming, 2022, pp.1~8 (2022)[DOI]
- X. Wang, "Evolution and innovations in animation : a comprehensive review and future directions", Concurrency and Computation : Practice and Experience, 36(2) (2023)[DOI]
- P. Pouliou, "Speculative hybrids: investigating the generation of conceptual architectural forms through the use of 3d generative adversarial networks", International Journal of Architectural Computing, 21(2), pp.315~336 (2023)[DOI]
- H. Du, "Generative ai-aided optimization for ai-generated content (aigc) services in edge networks" (2023)[DOI]
- N. Derevyanko, "Comparative analysis of neural networks midjourney, stable diffusion, and dall-e and ways of their implementation in the educational process of students of design specialities", Pedagogy and Psychology, 9(3), pp.36~44 (2023)[DOI]
- Y. Fan, "Full Face-and-Head 3D Model With Photorealistic Texture", IEEE Access, 8, pp.210709~210721 (2020)[DOI]
- S. Hörmann, "Face Texture Generation And Identity-Preserving Rectification", 2021 IEEE International Conference on Image Processing (ICIP), pp.2448~2452 (2021)
- K. Ma, "A Research on 3D Texture Production Using Artificial Intelligence Softwear", International Journal of Internet, Broadcasting and Communication, 15(4), pp.178~184 (2023)[DOI]
- W. Chujitarom, "The development of an artificial intelligence artist assistant(aiaa)model for the purpose of innovative digital storytelling in digital art education", Tem Journal, pp.1133~1140 (2024)[DOI]