MA-INF 2213 - Advanced Computer Vision

3L + 1E, SS25

Lecturer

Juergen Gall

Content

Tentative: linear methods for classification and regression, Gaussian processes, random forests, SVMs and kernels, convolutional neural networks, vision transformer, generative adversarial networks, diffusion models, vision-language models, structured learning, image classification, object detection, action recognition, pose estimation, face analysis, tracking, image synthesis.

Material

The slides and recordings will be provided via sciebo: Slides/Recordings.

Prerequisites

MA-INF2201 is recommended but not required.

Schedule

Wednesday, 10:15-11:45, CP1-HSZ / Hörsaal 3

Friday, 10:15-11:45, CP1-HSZ / Hörsaal 3 (every second week)

Start: Wednesday, 9.04.

Exercises

Theory and programming. At least 50% of the exercise points are required to qualify for exam. The programming exercises are implemented in Python and OpenCV.

Register for the exercises until 10.04. via the registration form

Schedule

Friday, 10:15-11:45, CP1-HSZ / Hörsaal 3 (every second week; sometimes on Wednesday)

Exam

Registration period: 01.06.-21.06.

Exam: 28.7.-29.7., 2.037, Informatikzentrum (tbc)

2nd Exam: 8.9.-9.9., 2.037, Informatikzentrum (tbc)