Parallel Epipolar Descriptor Correspondence
by Hunter, Quinn T., M.S., UNIVERSITY OF CALIFORNIA, DAVIS, 2011, 36 pages; 1507153

Abstract:

In order to properly triangulate three-dimensional geometry from the correspondence of two-dimensional image plane points, I created a GPU search algorithm for matching pixels in the image plane. All integer pixels in an image set to be reconstructed are tagged with multi-color DAISY descriptor vectors and set into memory. I calculate epipolar lines in each image from the epipolar planes, such that every integer pixel is covered in a particular image whose depth is to be projected out. The pixel descriptors are then rearranged in memory to match the format of the epipolar geometry. I then execute GPU kernels to solve pixel matches with threads grouped by lines created by a particular epipolar plane. Using a GPU in this fashion achieves an at least 4x speed-up from a CPU calculation, and another 4x speed-up is achieved using a coarse-to-fine image method in the same epipolar arrangement mentioned.

 
AdviserKenneth I. Joy
SchoolUNIVERSITY OF CALIFORNIA, DAVIS
SourceMAI/ 50-05, p. , Mar 2012
Source TypeThesis
SubjectsElectrical engineering; Computer science
Publication Number1507153
Adobe PDF Access the complete dissertation:
 

» Find an electronic copy at your library.
  Use the link below to access a full citation record of this graduate work:
  http://gateway.proquest.com/openurl%3furl_ver=Z39.88-2004%26res_dat=xri:pqdiss%26rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation%26rft_dat=xri:pqdiss:1507153
  If your library subscribes to the ProQuest Dissertations & Theses (PQDT) database, you may be entitled to a free electronic version of this graduate work. If not, you will have the option to purchase one, and access a 24 page preview for free (if available).

About ProQuest Dissertations & Theses
With over 2.3 million records, the ProQuest Dissertations & Theses (PQDT) database is the most comprehensive collection of dissertations and theses in the world. It is the database of record for graduate research.

The database includes citations of graduate works ranging from the first U.S. dissertation, accepted in 1861, to those accepted as recently as last semester. Of the 2.3 million graduate works included in the database, ProQuest offers more than 1.9 million in full text formats. Of those, over 860,000 are available in PDF format. More than 60,000 dissertations and theses are added to the database each year.

If you have questions, please feel free to visit the ProQuest Web site - http://www.proquest.com - or call ProQuest Hotline Customer Support at 1-800-521-3042.