Stulp, F. (2001) Completion of Occluded Surfaces. Master's Thesis / Essay, Artificial Intelligence.
|
Text
AI_Ma_2001_FStulp.CV.pdf - Published Version Download (2MB) | Preview |
Abstract
The CAMERA-project is an EC-funded international network that seeks to acquire three dimensional models of industrial and historical buildings through analysis of digital images; the main focus being on range images. When taking digital images of scenes, it is hard to capture all the information contained in the real-world scene into the image. Often objects are blocking other objects from the sensors' view. These occluded objects are not fully visible in the image, and data needed for a complete 3D reconstruction is missing. In this dissertation, a method is presented for using information from the surroundings of an occlusion to hypothesize what would have been seen, had there not been an occlusion. This process is called completion. Cognitive aspects of this completion are discussed. The main part of this dissertation is devoted to a method of completing surfaces in range images, and finding a correct way of projecting texture onto these surfaces.
| Item Type: | Thesis (Master's Thesis / Essay) |
|---|---|
| Degree programme: | Artificial Intelligence |
| Thesis type: | Master's Thesis / Essay |
| Language: | English |
| Date Deposited: | 15 Feb 2018 07:30 |
| Last Modified: | 15 Feb 2018 07:30 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/9061 |
Actions (login required)
 |
View Item |