Video rate textured image generation with anti-aliasing enhancements
Video rate textured image generation with anti-aliasing enhancements
This thesis explores several issues in the field of video-rate texture mapped imagery generation. Two key areas in particular are examined, these being the development of a cost-effective image generator, which has potential both as a high-performance system and as a test bed for future research, and the enhancement of the subjective image quality by suppression of the aliasing artifacts.
The development of a video rate textured image generator is first presented. This system uses the ray tracing image generation method for evaluating the mapping between screen and texture spaces. The ray tracing algorithm is organised to generate pixels in a raster scan order and is synchronized to the screen refresh signals which results in the elimination of the image framestore. Images are generated at a high refresh rate, in order to minimise the overall response time of the system.
The process by which the subjective quality of the textured images is enhanced is considered next. A spatially-variant method is devised which suppresses aliasing by progressively reducing the contrast of the image at the points where artifacts will occur. This method takes advantage of previously computed results in the image generator in order to determine the points in the image where aliasing suppression should be applied. This method does not require extra texture memory to store pre-computed results as is the case with other more conventional approaches.
In addition to the other areas examined, the use of an alternative colourspace for texture representation is investigated. The use of the YUV colourspace, as used in television broadcasting systems, in conjunction with a novel colour value encoding method is shown to reduce texture memory requirements as well as simplifying processing.
University of Southampton
1994
Carter, Matthew Joseph
(1994)
Video rate textured image generation with anti-aliasing enhancements.
University of Southampton, Doctoral Thesis.
Record type:
Thesis
(Doctoral)
Abstract
This thesis explores several issues in the field of video-rate texture mapped imagery generation. Two key areas in particular are examined, these being the development of a cost-effective image generator, which has potential both as a high-performance system and as a test bed for future research, and the enhancement of the subjective image quality by suppression of the aliasing artifacts.
The development of a video rate textured image generator is first presented. This system uses the ray tracing image generation method for evaluating the mapping between screen and texture spaces. The ray tracing algorithm is organised to generate pixels in a raster scan order and is synchronized to the screen refresh signals which results in the elimination of the image framestore. Images are generated at a high refresh rate, in order to minimise the overall response time of the system.
The process by which the subjective quality of the textured images is enhanced is considered next. A spatially-variant method is devised which suppresses aliasing by progressively reducing the contrast of the image at the points where artifacts will occur. This method takes advantage of previously computed results in the image generator in order to determine the points in the image where aliasing suppression should be applied. This method does not require extra texture memory to store pre-computed results as is the case with other more conventional approaches.
In addition to the other areas examined, the use of an alternative colourspace for texture representation is investigated. The use of the YUV colourspace, as used in television broadcasting systems, in conjunction with a novel colour value encoding method is shown to reduce texture memory requirements as well as simplifying processing.
This record has no associated files available for download.
More information
Published date: 1994
Identifiers
Local EPrints ID: 458514
URI: http://eprints.soton.ac.uk/id/eprint/458514
PURE UUID: 982c2f20-22f1-46ef-9edf-43bfc6c120f4
Catalogue record
Date deposited: 04 Jul 2022 16:50
Last modified: 04 Jul 2022 16:50
Export record
Contributors
Author:
Matthew Joseph Carter
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics