Computer Vision

Prof. Thomas Brox

Computer vision is a very active research field with many practical applications, for instance in quality control, robotics, driver assistance systems, and many more. The ultimate goal of computer vision is to imitate the great capabilies of the human visual system, allowing the computer not only to record images but also to interpret them. Significant progress has been made in recent years. This course introduces the most important concepts in today's Computer Vision research. The exercises will consist of programming assignments in Python, where you will learn to implement and/or test the most important techniques presented in the lectures.

The lecture was planned to be held partially online and partially in presence. Due to a too small room and far more students being interested in attending the lecture life than in earlier years, in addition to the pandemics getting again more severe in Germany, it must stay online. If you want to connect with other students in presence, you can use one of the rooms in building 101 and watch the online lecture from there. A group of students planned to meet before the lecture in the main hall of the building. Please check the procedure for using these rooms. Particularly, due to the critical situation in hospitals, you must be vaccinated to use these rooms and you must wear a mask.

Lecture: (2 SWS)	Tuesday, 10:15-11:45, Room: 52-2-17
Exercises: (2 SWS)	Thursday, 13:00-14:00 (first session October 28, 2021), Zoom link Contact persons: Jan Bechtold, Osama Makansi
Beginning:	Tuesday, October 19, 2021 via Zoom You must be registered for the course in HisInOne by Oct. 17 to receive the meeting password.
ECTS Credits:	6
Recommended semester:	2 or 3 (Master)
Requirements:	Fundamental mathematical knowledge and programming skills in Python. Image Processing and Computer Graphics teaches important technical knowledge that won't be repeated in Computer Vision. Thus, prior attendance of Image Processing is highly recommended. The online meetings will NOT be recorded.
Language:	The lectures are given in English.
Exam:	Written exam after the lecturing period.

Material:

Lecture material will be updated during the course. Some of the recordings are a bit outdated. Recording files are about 200MB each. *denotes lectures with an exercise.

Class 1 (19.10.):	Introduction	Recordings	Online lecture
Class 2 (26.10.)*:	Diffusion filters, TV minimization	Recordings	Online lecture
Class 3 (2.11.):	Image segmentation and combinatorial optimization	Recordings	Online lecture
Class 4 (9.11.):	Spectral clustering	Recordings	Online lecture
Class 5 (23.11.)*:	Deep learning I	Recordings	Online lecture
Class 6 (30.11.)*:	Deep learning II	Recordings	Online lecture
Class 7 (7.12.):	Optical Flow I	Recordings	Online lecture
Class 8 (14.12.)*:	Optical Flow II	Recordings	Online lecture
Class 9 (21.12.):	Video segmentation	Recordings	Online lecture
Class 10 (11.1.)*:	3D geometry and camera calibration	Recordings	Online lecture
Class 11 (18.1.)*:	Stereo reconstruction	Recordings	Online lecture
Class 12 (25.1.)*:	Disparity estimation and scene flow	Recordings	Online lecture
Class 13 (1.2.)*:	Structure from motion	Recordings	Online lecture
Class 14 (8.2.):	Deep learning III	Recordings	Online lecture

Exercise material:

Short introduction to git
Setup instructions
Exercises on github