Courses in Winter Term

Here you can find an overview of our group’s courses given during Winter terms. You can click on each course to find out more detailed information about it.


Rechnersehen 1

for B.Sc. and M. Sc. studies

This course is the first half of the Rechnersehen lecture block. It covers essential topics in the area of image processing with a focus on techniques related to signal processing. The course starts with image structures and the image acquisition process. After that, the theory of linear systems is introduced to model noise processes and introduce techniques to counteract them using filtering. Additionally, the theory of Fourier analysis and transformation is introduced to manipulate images in the frequency domain. In the next part, image features and their extraction are covered, as well as segmentation of lines, regions and textures. The lecture is accompanied by a practical exercise. There is no prior knowledge in computer vision necessary, but you should have finished most of the basic mathematics courses (Linear Algebra, Calculus, Numerics, Stochastic) of the Computer Science study program before attending this course.

Einführung in Tiefe Lernverfahren

for B. Sc. and M.Sc. studies

This is a course on deep learning covering basics, methods and applications. It also covers machine learning basics and provides useful tips for implementing deep learning techniques. In more detail, the topics of this course include: history and background, mathematical basics, training techniques, convolutional and recursive neural networks, popular software libraries and current research topics. To attend this course, no prior knowledge is necessary, but finishing Mustererkennung first is useful.

Maschinelles Lernen und Datamining

M. Sc. studies

This advanced course focuses on learning concepts with applications in data mining. In the first part the concept of machine learning is introduced. After that, the course covers techniques for data pre-processing (scales, relations, outlier) and visualization (PCA, MDS, ICA, etc.). The next part introduces different classification concepts like Bayes rule, log-linear models or decision trees. Very relevant for data exploration is also the chapter on grouping concepts like hierarchical, relational and spectral clustering. The last part covers dependency modelling using correlation, association, as well as Markov and Bayes nets. We highly recommend to finish the course Mustererkennung, before attending this lecture.

Zustandsschätzung und Aktionsauswahl

for M. Sc. studies

This advanced course focuses on two important aspects of sensor data processing: state estimation based on (noisy) observations, and optimal selection of actions given these (defective) estimations. In the first part of the course, classic techniques for state estimation of deterministic and stochastic systems are introduced (e.g. Kalman Filter and Particle Filter). The second part focuses on methods to directly influence the acquisition process to retrieve new sensor data. Based on Markov models the concept of reinforcement learning is introduced. As an alternative for action selection an information theory-based method based on MMI is shown. In the last part, this course also covers topics related to sensor data fusion. This lecture is accompanied by practical exercise. We highly recommend to finish all basic mathematics courses (especially Calculus and Stochastics) and at least one basic computer vision course like Rechnersehen 1 or Visuelle Objekterkennung.

Signalorientierte Bildverarbeitung

for M. Sc. studies

This advanced course focuses on the signal processing aspects of digital image processing. It mainly covers transformation methods like Fourier, Hadamard, Hilbert and Wavelets. For each of those methods practical examples with respect to image processing are shown. Typical important tasks are sampling, as well as image registration, compression and restoration. We highly recommend to finish Rechnersehen 1 first, before attending the lecture.

Spezielle Probleme im Rechnersehen

for M. Sc. studies

This is a very research-ortiented course for topics in the field of computer vision. Contents of this course highly vary and cover most recent research of the Computer Vision Group Jena. Students, PhD students and PostDocs of the group present their own research topics in talks, which are then discussed. As a prerequisite for admission in this course, you have to do a project work, Bachelor thesis or Master thesis in our group at the same time. Regular attendance is required, as well.

Informatik (B.Sc. Werkstoffwissenschaften)

for B. Sc. studies

This is an entry level course on computer science for students of B.Sc. Material Science. Given a pre-selected programming language (e.g. C++ or Python), this course covers mainly basics in programming. Additionally, there are chapters on basic computer science knowledge like algorithms, data structures, recursion and complexity. This lecture is accompanied by practical programming exercise.

Informatik II (B.Sc. Physik)

for B. Sc. studies

This is the second part of the entry level course on computer science for students of B.Sc. Physics. Following up on topics of the previous course Informatik I, this lecture covers object-oriented programming, as well as computer science basics on modular programming, pointers, exception handling, GUIs and computer networks. This lecture is accompanied by practical programming exercise.


Rechnersehen / Fortgeschrittene Methoden im Rechnersehen

for B. Sc. and M. Sc. studies

This course covers advanced methods and techniques in computer vision in greater depth. Topics are not fixed, but rather align with recent state-of-the-art literature in various areas of computer vision research. Typical areas for topics are deep learning, supervised and unsupervised learning, reinforcement learning, 3D vision, as well as application works in e.g. biodiversity research, environmental science, medical research or autonomous driving. In seminars, students will have to work on a certain topic on their own. At the end of the term they need to submit a term paper (essay) and give an oral presentation of their work.


Werkzeuge der Mustererkennung und des Maschinellen Lernens

for B. Sc. and M. Sc. studies

This is a hands-on course in which students can apply the techniques they have learned in Mustererkennung on real-world data. It covers all steps of a pattern recognition pipeline starting with data pre-processing up to the actual data analysis. The main focus is practical work in small groups to tackle weekly tasks. Experienced researchers provide guidance and give insight into each of the required steps. We highly recommend to finish the course Mustererkennung, before attending this lecture.

Anwendungspraktikum 3-D Rechnersehen / Projekt “Intelligente Systeme”

for B. Sc. and M. Sc. studies

In this course students work independently in small teams on specific computer vision software projects. The goal is to familiarize yourself with image processing and machine learning algorithms in a hands-on manner. Students can propose own projects, but there is also always a set of pre-defined topics to choose from. The course is usually accompanied by some small lectures on tools (e.g. Git, Docker, Python, etc.) and scientific methods. At the end of the project, students need to submit a documentation of their work. During the term, there are milestone meetings with the supervisor. We highly recommend to finish at least courses Rechnersehen 1 and/or Rechnersehen 2 before attending this course.