InViTe: Individual Virtual Transfer for Personalized 3D Face Generation System
InViTe: Individual Virtual Transfer for Personalized 3D Face Generation System
Mingyu Jang, Kyungjune Lee, Seongmin Lee, Hoseok Tong, Juwan Chung, Yusung Ro, Sanghoon Lee
Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence
Demo Track. Pages 8683-8686.
https://doi.org/10.24963/ijcai.2024/1008
With the expansion of the virtual communication industry using VR/AR, it has attracted increasing attention to enable users to represent their personalities in a 3D avatar. As the face of 3D avatars plays a crucial role in conveying human personality, a system that generates and manipulates 3D faces is desired. However, establishing the system is challenging due to the need for human effort and specialized knowledge. To fill this void, we present the Individual Virtual Transfer (InViTe), which enables the creation and customization of a 3D face according to the user's preference. Our proposed system is featured for 1) 3D face reconstruction with high fidelity texture map, 2) 3D face personalization, 3) realistic rendering results, and 4) real-time mobile virtual applications. We conduct an experiment to demonstrate that the proposed system can achieve sufficient individual personalization of 3D faces. Furthermore, we evaluate the system's data transmission protocol and demonstrate its efficiency. The demonstration video is available at https://www.youtube.com/watch?v=D 4pXZvGUWU.
Keywords:
Computer Vision: CV: 3D computer vision
Computer Vision: CV: Applications