About CS667
Professor:
Steve Marschner, srm@cs.cornell.edu
Office hours: M 2:30-3:30; W 1:30-2:30
Admin:
Kelly Patwell, patwell@cs.cornell.edu
Time and place:
Tuesday and Thursday, 2:55–4:10, 145 Olin Hall
Textbook:
Advanced Global Illumination, Second Edition, by Dutré, Bala, and
Bekaert. A K Peters, 2006.
Overview
This class covers that part of computer graphics, and especially rendering, that connects to physical models of the real world. The main problem areas we will look at are light transport and light reflection, but this year we will also discuss some ideas from computational photography—a set of techniques for measuring the real world for graphics and imaging applications.
Light transport is the fundamental rendering problem of modeling how light flows around in an environment and determining the equilibrium state that becomes a rendered image. Light reflection forms the boundary conditions for light transport: the challenge is to build practical computer models for how light interacts with the surfaces in the environment. As we look at these problems in this class we'll focus on getting to some traditionally “advanced” topics in scattering: participating media and translucent materials.
Coursework
The coursework consists of:
- solving paper-and-pencil problems for most lecture topics,
- implementing programming assigments in a physically based renderer,
- reading various articles and papers and giving me brief summaries and discussion questions,
- presenting one paper from the computer graphics literature to the class, and
- proposing, developing, and implementing a final project in teams of two to four.
I expect to have students with a variety of backgrounds, so there will be flexibility in the work, and grades will take each student's starting point into account. In particular, I no longer consider CS467 a prerequisite for this class; CS465 or another basic graphics class that covers ray tracing, or equivalent experience, should be sufficient. A certain level of mathematical sophistication, particularly in probability and multivariable calculus, will be required.
Steve Marschner (srm@cs.cornell.edu)