Piotr Dubla, PB and Tai Lucas, TG and Sebastian Murray-Rob, SL (2005) VizMark: Benchmarking Visibility Preprocessing, CS05-06-00, Department of Computer Science, University of Cape Town.
![[img]](http://pubs.cs.uct.ac.za/style/images/fileicons/application_pdf.png) |
PDF
tech_report.pdf Download (452kB) |
Abstract
We present a new means of comparing visibility algorithms by means of the implementation of a standard reference solution against which visibility algorithms may be tested. This will allow new and existing visibility algorithms to be objectively tested. An accurate reference solution was developed that employs an optimised ray casting algorithm to calculate visibility. Due to the excessive computational overhead in this calculations, a parallel implementation reduces the amount of time needed to produce the reference solution. The benchmarker component determines the accuracy of the algorithm undergoing testing based on a number of image error metrics which take into account the quality of the final, rendered image. This paper discusses the components that make up the VizMark system and how each is tested.
| Item Type: | Technical report |
|---|---|
| Uncontrolled Keywords: | visibility, ray casting, parallel computing, error metrics, rendered image quality |
| Subjects: | Computer systems organization > Architectures > Other architectures |
| Date Deposited: | 21 Oct 2005 |
| Last Modified: | 10 Oct 2019 15:35 |
| URI: | http://pubs.cs.uct.ac.za/id/eprint/243 |
Actions (login required)
 |
View Item |