摘要
针对约束纹理映射中三角网格极易发生形变的问题,本文提出一种基于均值坐标(Mean value coordinates)嵌入的约束纹理映射.使用基于局部/整体思想的ARAP(As-Rigid-As-Possible)算法参数化三维人脸网格,实验数据表明,参数化后的三维人脸网格三角形角度变形和面积变形均为最小.在网格嵌入中,采用均值坐标(Mean value coordinates)嵌入,得到的模型更加平滑.本文对多幅人脸图像进行纹理映射,结果表明该方法能够取得良好的实验效果.
This paper introduces a constrained texture mapping algorithm based on mean value coordinates embedding, sloving the geometric measurement distortion problem when texture mapping, which is expected to be as little as possible. Firstly, The algorithm adopts the as-rigid-as-possible algorithm to minimize the distortion of the angle and area of the triangle, then obtains the minimize deformation of the texture mapping. In addition, to reduce the degree of distortion in the grid embedded. This paper adopts the mean value coordinates to provide a smooth constrained mesh model. Using numerous photos as texture mapping, experimental results show that this method can obtain good result.
出处
《安徽师范大学学报（自然科学版）》
CAS
2016年第5期430-433,共4页
Journal of Anhui Normal University(Natural Science)
基金
国家自然基金(61371156)
安徽省科技攻关计划(140B042019)
芜湖市科技攻关项目(2016gg23)
关键词
纹理映射
ARAP参数化
均值坐标
texture mapping
ARAP parameterization
mean value coordinates
