- author = "Indriyati Atmosukarto",
- title = "{3D} Shape Analysis for Quantification, Classification, and Retrieval",
- school = "Computer Science and Engineering, University of Washington",
- year = "2010",
- address = "USA",
- keywords = "genetic algorithms, genetic programming",
-
URL = "
http://grail.cs.washington.edu/theses/AtmosukartoPhd.pdf",
- size = "139 pages",
-
abstract = "Three-dimensional objects are now commonly used in a
large number of applications including games,
mechanical engineering, archaeology, culture, and even
medicine. As a result, researchers have started to
investigate the use of 3D shape descriptors that aim to
encapsulate the important shape properties of the 3D
objects. This thesis presents new 3D shape
representation methodologies for quantification,
classification and retrieval tasks that are flexible
enough to be used in general applications, yet detailed
enough to be useful in medical craniofacial
dysmorphology studies. The methodologies begin by
computing low-level features at each point of the 3D
mesh and aggregating the features into histograms over
mesh neighbourhoods. Two different methodologies are
defined. The first methodology begins by learning the
characteristics of salient point histograms for each
particular application, and represents the points in a
2D spatial map based on longitude-latitude
transformation. The second methodology represents the
3D objects by using the global 2D histogram of the
azimuth-elevation angles of the surface normals of the
points on the 3D objects.
Four datasets, two craniofacial datasets and two general 3D object datasets, were obtained to develop and test the different shape analysis methods developed in this thesis. Each dataset has different shape characteristics that help explore the different properties of the methodologies. Experimental results on classifying the craniofacial datasets show that our methodologies achieve higher classification accuracy than medical experts and existing state-of-the-art 3D descriptors. Retrieval and classification results using the general 3D objects show that our methodologies are comparable to existing view-based and feature-based descriptors and outperform these descriptors in some cases. Our methodology can also be used to speed up the most powerful general 3D object descriptor to date.",
- notes = "GPLAB, Matlab",
