DM-301-2 Computer Graphics

Required optional subject in the master program Digital Media (6 Credits)
If you attend this course or plan to do so, please send me an e-mail so that I can compile a mailing list.

Schedule

Tuesday, 9:45 - 11:15 a.m. and 11:30 a.m. - 1 p.m., room I-117 or on announcement in the computer graphics lab I-220

Intended Learning Outcomes

The students are aware of the current topics and trends in research and development in the field of computer graphics, in particular concerning 3D reconstruction and image-based rendering. They are able to implement typical algorithms found in current scientific and/or technical papers that are based on intermediate-level methods of mathematics and physics. They conceive of ideas how to promote the state of the art in the science and technology of computer graphics and can underpin these ideas through literature research and practical experiments. They can appropriately present their work and their ideas to a scientific and/or technical audience in speech, writing, and images.

Prerequisites

Mathematics: linear algebra, calculus of one variable
Computer Science: object-oriented programming; algorithms and data structures
Computer Graphics: basics of modeling, animation, and rendering; software development using a real-time 3D application programming interface

If you lack sufficient background in computer graphics, I strongly recommend that you attend my introductory computer graphics course, too.

Coursework and Assessment (Tentative)

I'd like to see the following outputs of this course and base the grades solely on them:

design and implementation of one of the following:

a standard approach to some computer graphics problem (chosen on arrangement with me) in a robust and convenient form that addresses a large audience, for instance as a plug-in for off-the-shelf graphics software
a novel approach to some computer graphics problem (chosen on arrangement with me) as a proof of concept

presentation of the work as a scientific poster, accompanied by an extended abstract (possibly to be submitted to a computer graphics conference)

Group work of up to three people is ok if the task justifies it. I'll examine the individual contribution of every member of a group.

The intended workload (presence plus work at home or in the lab) is 6 credits times 30 hours per credit = 180 hours.

I recommend that you collect information from everywhere. But I expect everybody to understand the material he or she submits. Furthermore, any material that you did not write, draw, etc. on your own has to be clearly declared as such.

In the context of this course, I regard the poster as the documentation required by the examination regulations.

I'll grade the coursework according to the following table:

Grade Solution of Problem (50%) Structure (20%) Naming (5%) Comments (5%) Poster (20%)

5.0 vital part missing spaghetti muddled virtually no comments important part(s) missing in the poster or severe formal problems

4.0 mostly complete, but for instance erroneous behavior in important exceptional cases some logical problems, no defensive programming partially confusing every class and every non-trivial method explained with comments that are automatically extractable (in the context of C# that means XML) poster mostly complete, but for instance too little graphical material or significant gaps in references or explanations

3.0 complete up to minor gaps basically object-oriented, basic defensive programming comprehensible in addition: explanations of code that cannot be understood on first linear reading (If there are no difficult-to-read passages in the code, that's even better!); do not comment every single line poster virtually complete

2.0 complete clean classes built using design patterns; defensive programming; instrumented code self-explaining in addition: comments describing vital algorithms in brief poster and extended abstract ready for [hypothetical] submission as a poster to SIGGRAPH 2005, SCA 2005, or similar

1.0 all of the above (grade 2.0) plus significant further achievements such as a novel algorithm or a user interface of commercial quality

Grade	Solution of Problem (50%)	Structure (20%)	Naming (5%)	Comments (5%)	Poster (20%)
5.0	vital part missing	spaghetti	muddled	virtually no comments	important part(s) missing in the poster or severe formal problems
4.0	mostly complete, but for instance erroneous behavior in important exceptional cases	some logical problems, no defensive programming	partially confusing	every class and every non-trivial method explained with comments that are automatically extractable (in the context of C# that means XML)	poster mostly complete, but for instance too little graphical material or significant gaps in references or explanations
3.0	complete up to minor gaps	basically object-oriented, basic defensive programming	comprehensible	in addition: explanations of code that cannot be understood on first linear reading (If there are no difficult-to-read passages in the code, that's even better!); do not comment every single line	poster virtually complete
2.0	complete	clean classes built using design patterns; defensive programming; instrumented code	self-explaining	in addition: comments describing vital algorithms in brief	poster and extended abstract ready for [hypothetical] submission as a poster to SIGGRAPH 2005, SCA 2005, or similar
1.0	all of the above (grade 2.0) plus significant further achievements such as a novel algorithm or a user interface of commercial quality

DM-301-2 Computer Graphics

Schedule

Intended Learning Outcomes

Prerequisites

Recommended Reading

Coursework and Assessment (Tentative)