Overview

In this seminar we will discuss the diverse set of paradigms for generative modeling in the area of computer vision. We will cover both seminal works, such as diffusion models, flow matching, and sequence generation paradigms, as well as recent advances in generative modeling for solving inverse problems in 2D and 3D computer vision, such as conditional object and scene generation/reconstruction, novel view synthesis, inpainting, and image super-resolution.
 

The seminar will consist of an introductory meeting with a lecture at the beginning of the semester introducing the field and distributing papers, and a two-day block course in the semester break covering paper presentations and discussions. Students are expected to read into their assigned paper, the related literature, prepare a talk as well as a paper summary with critical discussion.

Course Information

Semester:  SS

Year:  2025

Requirements: The student has a solid understanding of of Machine Learning, Computer Vision and feels comfortable with Neural Networks (for example through lectures High Level Computer Vision, Neural Networks: Theory and Implementation, or Machine Learning).

Time and location:

Introductory lecture:  06.05.25  10:00 - 12:00    E1.4 room 021 (MPII seminar room)

Block seminar:  13.08.25  &  14.08.25  10:00 - 16:00   E1.4 room 021 (MPII seminar room)

Essay is due on 31.07.25

Registration:   https://seminars.cs.uni-saarland.de/sose25seminars

Lecturer:           Dr. Jan Eric Lenssen, Mohammad Asim

Material

Intruductionary slides and paper assignments will appear hear as soon as the Seminar has started.