基于二维视图的三维人脸重建设计  

Design of 3D face reconstruction based on 2D view

作　　者：杨静[1] YANG Jing(School of Information Engineering,Anhui Business and Technology College,Hefei 231131,China)

机构地区：[1]安徽工商职业学院信息工程学院,安徽合肥231131

出　　处：《商丘师范学院学报》2024年第12期20-24,共5页Journal of Shangqiu Normal University

基　　金：安徽省高校自然科学研究项目(KJ2021A1511)。

摘　　要：人脸表情是交流展示的重要信息载体,三维人脸重建在图像识别、动画影视等领域有至关重要的作用.针对三维人脸重建存在细节易丢失、计算复杂度高等问题,提出基于二维视图的三维人脸重建设计.以二维视图为基础,利用傅立叶神经算子来求解泊松方程,并从定向点云测量中重建网格,改进经典深度神经网络在形状重建中的局限性,在重建质量、运行时间等方面取得较好结果.此外,为量化人脸图像表情识别,提出基于顶点距离的测量方法,结合支持向量机对表情状态进行分类.Facial expression is an important information carrier for communication and display,and 3D face reconstruction plays a vital role in image recognition,animation,film and other fields.Aiming at the problems of losing details and high computational complexity in 3D face reconstruction,a 3D face reconstruction design based on 2D view is proposed.Based on two-dimensional view,the method proposed in this paper uses Fourier neural operator to solve Poisson equation,and reconstructs the mesh from directional point cloud measurement,which improves the limitation of classical depth neural network in shape reconstruction and obtains good results in reconstruction quality and running time.In addition,in order to quantify facial expression recognition,a measurement method based on vertex distance is proposed,and the expression state is classified by support vector machine.

关 键 词：三维人脸重建 二维视图 傅立叶神经算子 顶点距离 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]