Direkt zum Inhalt springen
Computer Vision Group
TUM Department of Informatics
Technical University of Munich

Technical University of Munich



Practical Course: Correspondence and Matching Problems in Computer Vision (10 ECTS)

Winter Semester 2020/2021, TU München

Organisers: Prof. Dr. Florian Bernard, Maolin Gao, Tarun Yenamandra

Please direct questions to cmpcv-ws20@vision.in.tum.de


2020-09-11: [gitlab account] collect your gitlab using the table at the bottom of the page.

2020-08-18: [topic assignment] see the topic assignment and grouping here.

2020-08-14: [application closed] accepted applicants have been informed via mail.

2020-07-30: [application update] We have some vacancies open for applications. Please send your application (details see below) to cmpcv-ws20[at]vision.in.tum.de until 09 August 2020 23:59 CEST. Application results will be given till 14 August 2020 if successful.

2020-07-16: [application] Please send applications (indicating learning goals; grade transcripts; programming skills; background in mathematics, optimisation and visual computing) to cmpcv-ws20[at]vision.in.tum.de . Sending an email with sufficient info about yourself until 23 July is crucial for matching success.

2020-07-15: [preliminary meeting] Slides from the preliminary meeting are available here (you need to log-in with the same account that you use in TUMonline).

Course Description

Correspondence and matching problems are omnipresent in computer vision and related fields. Such problems can appear in many different contexts (e.g. image alignment, shape matching, object tracking, 3D reconstruction, etc.), which highlights their high relevance. Most commonly, such problems are phrased in terms of numerical optimisation problems, as machine learning prediction tasks, or as a combination of both. During this course students will work on a specific project that tackles a particular correspondence or matching problem setting. They will implement and analyse methods described in existing research papers, as well as understand and analyse their strengths and weaknesses. The topics will be related to:

  • Graph matching
  • Image alignment
  • Point-cloud registration
  • Shape matching
  • Deep learning
  • Feature learning
  • Multi-matching problems
  • Cycle-consistency of maps

It is expected that all participants have excellent programming skills in a scientific programming language (e.g. Matlab or Python), a thorough understanding of algorithms and data structures, as well as a solid working knowledge of linear algebra and calculus.

In addition, students should have a background in at least one of the following topics: continuous/discrete optimisation, 3D geometry, computer vision, image processing, or computer graphics.

Moreover, important soft skills include communication skills, the ability to identify what is unclear, to figure out what questions need to be asked to clarify it, to formulate the questions clearly, and to ask the tutor without hesitation. Communicating well and strategically is an important rule of the practical course. Almost all difficulties experienced by students are due to not following these rules.

Course Structure

At the beginning of the course, relevant basics of correspondence and matching problems, including theoretical and practical aspects, will be covered in lectures. Subsequently, students will work on practical projects in groups of 2-3. In order to ensure an appropriate progress of the students, there will be regular communication with tutors. At the end of the project, each group will present their project with a following Q&A session. There will be no additional written or oral exam. Both the theoretical and practical part of the project will be considered in the final grading.

Course Schedule

See here.

GitLab Account

Rechte Seite

Informatik IX
Chair of Computer Vision & Artificial Intelligence

Boltzmannstrasse 3
85748 Garching info@vision.in.tum.de

Follow us on:
CVG Group DVL Group



Bernt Schiele (Max Planck Institute for Informatics) will give a talk in the TUM AI lecture series on June 10th, 3pm! Livestream

French-German Machine Learning Symposium

French-German Machine Learning Symposium

The French-German Machine Learning Symposium aims to strengthen interactions and inspire collaborations between both countries. We invited some of the leading ML researchers from France and Germany to this two-day symposium to give a glimpse into their research, and engage in discussions on the future of machine learning and how to strengthen research collaborations in ML between France and Germany.

The list of speakers includes Yann LeCun, Cordelia Schmid, Jean-Bernard Lasserre, Bernhard Schölkopf, and many more! For the full program please visit the webpage.


Ron Kimmel (Technion - Israel Institute of Technology) will give a talk in the TUM AI lecture series on May 6th, 3pm! Livestream


4Seasons Dataset: We have released a novel dataset for benchmarking multi-weather SLAM in autonomous driving.


Hao Li (Pinscreen) will give a talk in the TUM AI lecture series on April 22nd, 8pm! Livestream