Reading #1: problem from chapter 8 - Supersampling in Ray Tracing


the problem: In generating three-dimensional computer graphics with ray-tracing, it is crucial to determine as accuratly as possible exactly how much light each pixel receives from the various light sources and reflections in the scene. This is accomplished by shooting more than one ray through each pixel, however, the exact points through which these rays travel are generated randomly for each pixel. As such, it is possible for confusions to arise about how the objects visible in each pixel are arranged. If there is a large discrepency in the placement of the objects, then that particular set of rays is no good and a new set has to be generated. The problem is to come up with an algorithm that determines the discrepency in each set of points.
terms: pixels: the tiny dots composing the images on a computer screen

rendering: the process of generating a 3-D scene

ray-tracing: a rendering method that operates by shooting rays through the scene to determine which objects and how much light is visible in each pixel.

supersampling: in ray-tracing, using many sample points per pixel.

keywords: ray-tracing

supersampling

web search: ray-tracing bibliography

Ray Tracing News

POV Access (search engine/links on ray-tracing)

A paper on a data distributed, parallel algorithm for ray-traced volume rendering

a page on supersampling from a stanford computer science class
biblio search: At Max-Planck-Institut fur Informatik I found:
H. Mučller: Raytracing complex scenes by grids.
Report ??, Inst. Inform., Univ. Karlsruhe, 1985.

At Universiteit Utrecht I found:

@techreport{ff-rtiph-85
, author =	"E. Fiume and A. Fournier"
, title =	"Ray-tracing is {$PSPACE$}-hard"
, type =	"Report"
, number =	"??"
, institution =	"Dept. Comput. Sci., Univ. Toronto"
, address =	"Toronto, ON"
, year =	1985
}

@techreport{m-rcsg-85
, author =	"H. M{\"u}ller"
, title =	"Raytracing complex scenes by grids"
, type =	"Report"
, number =	"??"
, institution =	"Inst. Inform., Univ. Karlsruhe"
, address =	"Karlsruhe, West Germany"
, year =	1985
}

@techreport{nt-abtrt-86
, author =	"Bruce Naylor and William Thibault"
, title =	"Application of {BSP} Trees to Ray-Tracing and {CSG}
Evaluation"
, type =	"Technical Report"
, number =	"GIT-ICS 86/03"
, institution =	"Georgia Institute of Tech., School of Information and
Computer Science"
, month =	feb
, year =	1986
, keywords =	"hierarchies"
, update =	"97.07 agarwal"
}



estimated time for this project:

skimming the chapters: half an hour
reading, note-taking: one hour
developing the homepage: one hour
typing the summary: one hour
web search: half an hour
bibliography search: twenty minutes

total time: 4 hours, 50 minutes


source for this assignment: Computational Geometry: Algorithms and Applications by M. de Berg et al. pp. 163-165