CASIA OpenIR  > 毕业生  > 硕士学位论文
Alternative TitleThree Dimension Face Reconstruction and Segmentation
Thesis Advisor马颂德
Degree Grantor中国科学院自动化研究所
Place of Conferral中国科学院自动化研究所
Degree Discipline模式识别与智能系统
Keyword三维计算机视觉 距离图像 三维重建 配准 非均匀有理b样条 控制点 三维部分分割 表面描述 3d Computer Vision Range Image 3d Reconstruction Nurbs Control Points 3d Part Segmentation Surface Description
Abstract三维表面形状重建在工程逆问题(由物体的图像或深度图,重建表面 形状,是典型的工程逆问题(Reverse Engineering),其研究具有重要 的理论意义)、工业设计、虚拟现实、三维传真、计算机辅助外科矫 形等领域有着重大的应用前景和实用价值。 这篇文章主要研究三维面像的形状恢复问题。与其它三维表面形状重 建工作不同,人的面像具有多种细微特征,是一个复杂的无规律表面。 首先,为了保证重建的精确度,我们自行设计了一个高精度激光扫描 系统采集距离图像。系统将人面部的彩色图像与三维深度图像(Depth Map)进行配准,以重建出高质量真实彩色的三维图像。然后,利用 非均匀有理B样条曲线(NURBS)的透射不变性,获取面部曲线的 控制点数据,拟合空间的面像线条。最后通过模拟电荷密度在导体表 面的分布,寻找局部极值点,搜索部分边界,将空间三维数据分割为 不同的部分,从而可以对每个部分进行表面描述。
Other AbstractThree dimension shape reconstruction is important for computing part-based descriptions and efficient object recognition. The development of fast and inexpensive 3D reconstruction systems opens up a vast range of applications, including: reverse engineering, industrial design, analysis and simulation, populating virtual environment and high speed 3D copy. This paper describes 3D face reconstruction and part segmentation. Compared with the previous 3D surface reconstruction, face is more complex, and face has many nice features and has no general rule on it. In order to describe it well, we try to describe its separate parts and then represent the face in terms of its parts and parts arrangement. First, We design a high quality laser scan system to acquire range data. By matching the color image with the range data, we get a photo-realistic 3D face image. Then, using the invariance of NURBS under perspective projection, we reconstruct the control points of the facial curves and fit the 3D facial curves. Last, based on a particular regularity in nature, "transversality", we simulate the electrical charge distributions on perfect conductor to find the local minimum and part boundary and segment 3D face into parts. Then we can get part description for a face.
Other Identifier559
Document Type学位论文
Recommended Citation
GB/T 7714
李红霞. 三维面像重建与分割[D]. 中国科学院自动化研究所. 中国科学院自动化研究所,2000.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[李红霞]'s Articles
Baidu academic
Similar articles in Baidu academic
[李红霞]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[李红霞]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.