Advanced Deep Learning for Computer Vision: Visual Computing

What it is about

The module conveys current deep-learning methods for computer vision. You will learn both theoretical foundations of modern neural architectures and practical methods such as generative models (GANs, diffusion models, Large Reconstruction Models) and modern 3D representations (e.g., Neural Radiance Fields, 3D Gaussian Splatting). In the end you will be able to apply, evaluate, and implement state-of-the-art methods in project work.

What you will be able to do

• Understand theoretical foundations of advanced network architectures
• Apply and evaluate current deep-learning methods for vision
• Translate research ideas into practical project work
• Analyze results and scientific communication

What the module consists of

• LectureConveying the theoretical foundations and current methods in computer vision
• Project work (group work in pairs)Application of lecture content to a concrete research problem; design, implementation and evaluation

Teaching method

• LectureExplanation of the theoretical foundations of modern networks and methods
• Project work in two-person teamsPractical application, experimentation, result analysis and scientific communication