One of the perennial goals of computer graphics is creating high quality images which are indistinguishable from photographs---a goal referred to as photorealism. Achieving this goal requires reproducing the appearance of natural materials, ranging from human faces and clothing to leaves, the sand and the sky. Historically, specific parametric lighting and reflectance models have been developed for specific phenomena. However, these approaches have been found wanting in describing the intricate complexity of real-world surfaces. In the last decade, a major revolution in realistic computer graphics has occurred, with it becoming more common to measure large datasets of appearance from the real world. These may include BRDFs, texture, or reflectance fields. Typically, these datasets allow very high visual fidelity, but are large and high dimensional. For example, to capture the variation at each surface location with lighting and view, we need a 6D table. Therefore, besides acquisition, it is critical to develop new representations suitable for data-driven visual appearance. These representations have had broad impact in computer graphics, ranging from real-time rendering to efficient appearance acquisition and simulation.
In this course, we will provide a taxonomy of various data-driven appearance representations, and review some of the recent ideas that seek to bridge the gap between realism, compression and interactivity. Topics include measured BRDF, texture and reflectance fields, spherical harmonic, wavelet and factored representations, and applications to real-time rendering, image-based rendering and compact material representation.
Below are some example images and computer renderings,
corresponding to the types of appearance we will be discussing.
We offer a full plate of computer graphics courses. I hope you will be enthusiastic about, and enrol in these offerings. The future offering of these courses will depend on your enthusiasm.
Specifically, the introductory course in computer graphics, COMS 4160 will be taught by me in the fall, and I presume most of you have taken it or an equivalent. In alternate years, I teach COMS 4162 which is a continuation course.
The course will consist of lectures on the relevant topics by the instructor, student presentations of papers covering current research in the area, and student projects. A syllabus/schedule is noted below. The grading will be 30% for paper presentations, 60% for the project, and 10% for class participation. In general, roughly (depends on the number of students in the course), 2 paper presentations will be required for those taking the course for a grade, and 1 for those taking the course pass/fail. A project is not required for students taking the course pass/fail. This is a good option for PhD students and others to read papers on this exciting topic and learn about the area without committing too much effort into a course project. Auditors, who simply want to sit in on the course are also welcome; however, we prefer if you sign up for the course pass/fail instead [this just involves doing one paper presentation]. For those of you who took COMS 6998 (appearance models in graphics and vision) in fall 2002, or high quality real-time rendering in fall 2004, the format of this course will be very similar. However, the content and focus are different, so you are welcome to repeat the course for credit.
Students taking the course for a letter grade are required to do a project [this may be in groups of 2-3], give a presentation in class regarding their results, and also submit a final written report. Wide flexibility is available with respect to project topics, provided they relate loosely to the subject matter of the course. We expect that most projects will implement one or more of the algorithms or papers discussed in the course. It is perhaps easiest to implement one of the papers that directly relates to real-time or offline rendering. However, other projects including acquisition, analysis or other topics are also welcome, and we very much appreciate suggestions from the students on alternative project ideas. The best projects will go beyond the published work in some way, such as trying out an alternative or better approach or trying to develop some variant or more general version of the technique. However, this is not essential; in general, students who fulfil all course requirements including a well-executed project will easily receive an A in the course.
As a potentially easier alternative to the project, we will also accept a well-written summary or tutorial, covering 3 or 4 papers. The best summaries will point out links between the papers not noticed by the original authors and suggest improvements or directions for future research. However, this option is recommended only as a last resort and will generally receive a slightly lower score; we prefer that you do a good project (which may involve understanding a few papers in any case).