Lecture Course: GPU Programming, Winter Term 2019/20

Students will obtain a good understanding of the graphics pipeline and its programmable stages. They will be able to distribute the tasks of a computer graphics algorithm among the different stages of the graphics pipeline efficiently. Moreover, they will be able to implement some of the most commonly used algorithms in real-time computer graphics. Furthermore, students will have a general idea of the GPU architectures and how to implement efficiently parallel algorithms in the GPU using Compute shaders and CUDA.

The lectures will cover in detail the different programmable stages of the OpenGL graphics pipeline: Vertex Shader, Fragment Shader, Geometry Shader, and Tessellation Shader. These programmable stages will be presented through a subset of the most commonly used computer graphics algorithms. The last part of the course will focus on the implementation of parallel algorithms on the GPU using both Compute Shaders and CUDA. The lectures and exercices will be in English.

The exercises will be interleaved with the given lectures.