Probabilistic Graphical Models and their Applications


This course will introduce the basic concepts of probabilistic graphical models. Graphical Models are a unified framework that allow to express complex probability distributions in a compact way. Many machine learning applications are tackled by the use of these models, in this course we will highlight the possibilities with computer vision applications.


The main goal of the class is to understand the concepts behind graphical models and to give hands-on knowledge such that one is able to design models for computer vision applications but also in other domains. Therefore the lecture is roughly divided in two parts: learning about graphical models and seeing them in action.


In the first part of the lecture we will discuss the basics of these models, eg. directed and undirected models, factor graph representations, learning of parameters, exact and approximate inference techniques. In the second part we will then discuss prominent applications for both low- and high-level computer vision problems. Some examples are statistical models of images (eg denoising), body pose estimation, person tracking, object detection and semantic image segmentation.


The exercises will be a mix of theoretical and practical assignments.

Course Information

Semester:  WS

Year:  2016/2017

Lecture start:  Wednesday, 26.10

Time and Location: 

Lecture: Wednesday 2pm - 4pm at MPI Informatik in room 024
Exercise: Thursday 10am - 12 pm at MPI Informatik in room 024

VL 2 + Ü 2

On 16.11. the lecture is in room 019.

Mailing List: send an email with your Matriculation Number to with [pgm-subscribe] in the subject

Lecturer(s):  Prof. Dr. Bernt Schiele 

                        Dr. Bjoern Andres (office hours: Fridays, 11am - 12pm)

TA(s): Eldar Insafutdinov (office hours: Tuesdays, 4 - 5pm)

            Evgeny Levinkov (office hours: Mondays, 4 - 5pm)


Exercises start Thu Nov 3nd


There is no exercise on Thursday, November 10


There is no lecture on Wednesday, November 16


There is no lecture on Wednesday, January 11 and a tutorial on Thursday, January 12


Two slots for oral exam available: March 8th and April 4th.


If you would like to register for an exam, please email Eldar (title: [PGM] Exam registration).



Main Textbook

  • Bayesian Reasoning and Machine Learning, David Barber, Cambridge University Press, online available at Textbook website (thanks David), we will discuss Chapters 1-6, a bit of 11, some parts of 27,28

Additional References

  • Probabilistic Graphical Models: Principles and Techniques, Daphne Koller and Nir Friedman, MIT Press, 2009 (careful: 1300 pages)
  • Pattern Recognition and Machine Learning, Chris Bishop, Springer, 2006