Computer Graphics: Conventional and Neural Methods

Here you can find all important information about the lecture and exercise of the course Computer Graphics: Conventional and Neural Methods (link to Zeus).

1. Lecture

Below you can find all information about the lecture.

1.1 Contents

An introduction to interactive computer graphics is given. This encompasses programming in WebGL. Participants will understand the way from the input data (geometric descriptions of objects) to the pixels of an output image. Steps are:

• Data processing
• Rasterization, Shading methods, Texturing
• Local vs. global illumination methods for photorealistic image generation
• Raytracing, radiosity and image-based rendering

In addition, machine learning methods for computer graphics will be discussed:  

• AI-generated images and video.
• Lightfields and NERF-based image generation
• procedural modeling, inverse modeling, non-tringular geometry.

The course will enable participants to use and modify neural networks for different applications in CG.

One main reference for the course is given in the following book:

Edward Angel, Dave Shreiner: Interactive Computer Graphics - a top down approach with WebGL (Chapters 1-9 and parts of 10-12 will be covered). The book is available as an electronic book in the library. Material for machine learning methods will be given out in the course.

1.2. Addressees

Students of computer science/information science

1.3. Subject area

Applied Computer Science

1.4. Offer for other fields of study

Mathematics, physics, minor or major in computer science in a master's degree program.

1.5. Prerequisites

Knowledge of JavaScript or another object-oriented programming language and a willingness to learn some JavaScript. The programming languages used are JavaScript as well as GLSL (OpenGL Shading Language). A brief introduction to WebGL and shader programming will be given in the tutorial.

It is necessary to have a computer capable of running at least WebGL 2 and an IDE like VSCode or similar.

Registration in Zeus.

1.6 Certificate of achievement

Written exam of 60 minutes duration at the end of the semester. Successful participation in the exercises is required for admission to the written exam.

1.7. Credits

With successful participation 6 credits can be earned.

1.8. Dates

First lecture: 26/10/2023 at 3:15 PM in ZT1202.

2. Exercise

Here you can find all information about the exercise (Zeus).

2.1 Contents

In the exercise you will program the learned content from the lecture.

2.2 Dates

Exercise 1: 30/10/2023 at 11:45 AM in ZT1202.
Exercise 2: 30/10/2023 at 3:15 PM in ZT1202.

2.3 Handing in the exercises

Exercises will be distributed and handed in via the university's Gitlab. You can log in with your university account. The submission of the exercises is described in detail in the Gitlab of the exercises (link visible after first exercise).

Questions?

Please write us an email: Patrick Paetzold or Michael Stroh.