Intelligent Facial Expression Generation Method Based on Artificial Intelligence

• Published in: Advances in Transdisciplinary Engineering
• Main Author: Wang K.; Chen J.; Bakar S.A.B.S.A.
• Format: Conference paper
• Language: English
• Published: IOS Press BV 2024
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85189539295&doi=10.3233%2fATDE231244&partnerID=40&md5=e67a42d2450dd640756ac48000c80348

This paper systematically analyzes and summarizes the related technologies of face modeling and expression animation based on artificial intelligence. Integrated with the implementation of graphics rendering engine, the system of face modeling and expression animation is designed and implemented first. Then, this paper adopts FAP and distance attenuation function to calculate the displacement of feature points and non-feature points, and realizes six kinds of basic expression modeling subsystems. With automatic calculation and manual operation, deep learning is used to design cascaded classifiers that extract facial expression feature images from real images and re optimize the 2D face corrected by the 3DMM model. The system of human face animation implementation is provided and it is proved to make further changes and extensions to the defined face model and face animation definition table, which can also meet various objective requirements, and it has the characteristics of smooth rendering and realistic animation effect. © 2024 The Authors.