Dissertation/Thesis Abstract

Piecewise surface reconstruction from range data
by Yu, Gene, Ph.D., City University of New York, 2010, 130; 3397418
Abstract (Summary)

Geometric modeling from range data is a long-standing problem in the field of computer graphics. It remains challenging even when the data is replete with simple surfaces such as planes and polynomials, as is common in urban scenes. The need to represent such scenes requires us to segment the data into its components. The collection of these components constitutes the final geometric model. This thesis introduces a method of piecewise surface reconstruction that fits a scene with a model composed of disjoint surfaces. The contribution of this work is the introduction of a surface evaluation method based on quantitative entropy measurements for balancing the tradeoff between accuracy and efficiency. Integrated surface evaluation enables us to produce output models that are accurate to within user-specified tolerances. Since our algorithm minimizes global criteria, it is robust to holes, occlusions, nonplanar surfaces, and missing data. Compared to methods that operate on unorganized point clouds and utilize no segmentation, our approach provides the user with greater control over the final appearance and error characteristics of the output model. A range of shape approximations such as plane, polynomial, and spline mesh surfaces can be used interchangeably. This flexibility is applicable to all scenes involving piecewise models.

Indexing (document details)
Advisor: Wolberg, George
Commitee: Grossberg, Michael, Reed, Michael, Stamos, Ioannis
School: City University of New York
Department: Computer Science
School Location: United States -- New York
Source: DAI-B 71/04, Dissertation Abstracts International
Subjects: Computer science
Keywords: Computer graphics, Lidar, Range segmentation, Surface reconstruction, Urban scenes
Publication Number: 3397418
ISBN: 9781109692655
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy