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

Technical University of Munich

Menu

Links

Informatik IX
Chair of Computer Vision & Artificial Intelligence

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

Follow us on:
CVG Group DVL Group

News

02.10.2020

We have five papers accepted to 3DV 2020!

30.09.2020

Our effcient deep network architectures form the AI engine of the project Slow Down COVID-19 at Harvard.

24.07.2020

Our practical course "Vision-based Navigation" (WS18, SS19) by Dr. Vladyslav Usenko and Nikolaus Demmel was honored as best practical course in the academic year 2018/2019 by the department for Informatics.

07.05.2020

We are organizing a workshop on Map-based Localization for Autonomous Driving at ECCV 2020, Glasgow, UK.

13.04.2020

Daniel Cremers received an ERC Advanced Grant (3.5 Mio Euro) for pioneering frontier research from the European Research Council. This constitutes his fifth ERC grant.

More



Lecture: Computer Vision II: Multiple View Geometry (IN2228)

SS 2020, TU München

News
  • 15/09:
  • 20/07:
    • Graded exercise FAQs are updated here
    • Relevant material: all lecture material (weeks 1-13) and the exercise material (week 1-12) are relevant for the final/retake
  • 08/07: We added clarifications on the graded exercise format, see graded exercise.
  • 07/07: The automatic registration change happened by mistake. If you want to participate in the mid-term but are now registered for the final, please manually unregister from the final and register for the mid-term (all students registered for the mid-term will automatically be registered for the final, as announced previously - this will happen after the mid-term).
  • 03/07: We are aware that the registration for the mid-term/final of some students have been changed in the TUMonline system - you will hear from us how to proceed within the next days.
  • 01/07: Registration for the final graded exercise, which will take place on July 28th 08:00-10:00, is open until July 13th. If you are registered for the mid-term until July 8th 23:59, you will automatically be registered for the final graded exercises, see Final Graded Exercise.
  • 22/06: Details on the (late) mid-term graded exercise are finalised.
  • 15/06:
    • The test run exercise will take place on July 16th (tentative).
    • The exam format will be a graded exercise, see below for details.
  • 20/04:
    • All materials relevant for this course can be found on this page (this lecture is not offered through Moodle).
    • For contacting us please use mvg-ss20@vision.in.tum.de, rather than using our personal mail addresses. It will ensure that you get a timely reply.
  • 03/04: At the time being it is likely that this lecture will be taught digitally only. We will dynamically adjust the format depending on the current developments regarding Covid-19. Please check this place regularly for updates.
Registration

If you plan to attend, please register for the course in TUMonline.
Later during the semester you will have to register for the exam.

Course Content

The lecture introduces the basic concepts of image formation - perspective projection and camera motion. The goal is to reconstruct the three-dimensional world and the camera motion from multiple images. To this end, one determines correspondences between points in various images and respective constraints that allow to compute motion and 3D structure. A particular emphasis of the lecture is on mathematical descriptions of rigid body motion and of perspective projection. For estimating camera motion and 3D geometry we will make use of both spectral methods and methods of nonlinear optimization.

Lecture

Pre-recorded lectures along with lecture slides will be made available weekly on Wednesday/Thursday.

Organisation: Prof. Dr. Florian Bernard
Contact: mvg-ss20@vision.in.tum.de
Start: Thursday, April 23, 2020

The lecture is offered in English.

Exam

Details on the exam can be found here.

Exercises

Exercises will be offered in a digital format. After each week's set of lectures, we will provide an assignment sheet that (usually) is a mixture of theoretical and practical exercises. Students are asked to independently work through these assignment sheets. The purpose of the exercises is to help students deepening their understanding of the material. They must not be submitted and there will not be any bonus for completed exercises. If there are any questions related to the exercises, students can ask them per email until Wednesday in the week after the assignment sheet is released. Subsequently, we will pre-record a video which covers solutions to the exercises and addresses questions raised by students. Since we will record these videos on Thursdays we can only include questions that reach us until Wednesday night. We plan to release the video on Friday.

Please note that we will not include questions about exercises other than the most recent one. If there are questions about earlier exercises, please contact us via email. In order that all students benefit from raised questions, we will select questions that we deem to be relevant for other students and provide the answer at our MVG Q&A page.

Organization: Mohammed Brahimi, Lukas Köstler
Contact: mvg-ss20@vision.in.tum.de
Start: Friday, April 30, 2020

Summary of Course Structure
  • Wed/Thu: We will release pre-recorded lecture videos. Students should study these as well as the provided lecture material and corresponding sections in the suggested literature (if necessary).
  • Thu: We will release an exercise sheet.
  • Thu-Wed: Students should work on the exercise sheet. Send any questions regarding this week's exercises that you'd like to be covered in our exercise-video to mvg-ss20@vision.in.tum.de.
  • Fri (starting from the second week): We will release an exercise-video and provide solutions to the exercise sheets. Students should study this material.

Note: of course students are welcome to study the provided material at their own pace. However, since in this case we cannot include answers in the exercise-video, we encourage students to adhere to our suggested schedule.

Inter-Student Communication

Our experience has shown that students significantly benefit from discussing and talking to their fellow students. Since without physically attending lectures and exercises it is difficult to meet fellow students, we have set up a student chatroom for discussions related to this course. Please note that this is not an official communication channel to reach the course organisers (please use email for this), but rather to provide a common space where students taking this course can meet each other and discuss things related to this course.

The student chat room uses the matrix protocol. To create a matrix account follow the instructions on the wiki: https://wiki.in.tum.de/Informatik/Helpdesk/RIOT . Make sure to create an account on exactly one server (either tum login xor in.tum login), as you will end up with multiple accounts otherwise. For joining the room, use https://matrix.in.tum.de/#/room/#mvg_student:tum.de if you have an account on the in.tum.de server (you log in with your in.tum account) or use https://matrix.tum.de/#/room/#mvg_student:tum.de if you have an account on the tum.de server (you log in with your tum account).

MATLAB

The programming exercises are done in MATLAB. You don't need extensive prior MATLAB knowledge. If you have never used MATLAB before, we recommend following a basic tutorial. Please install MATLAB on your laptop before the first exercise using the university's student licenses.

Note on MATLAB version:

We suggest you use a MATLAB version 2016b or later. For the solutions we make use of function definitions inside Matlab script files. This feature was added only in 2016b. So if you have an earlier version, we suggest you upgrade (alternatively you can move all function definitions to separate files). See also https://de.mathworks.com/help/matlab/matlab_prog/local-functions-in-scripts.html.

Lecture Material

Slides and exercises can be downloaded here. Registered students will receive an email with the password before the lecture period starts. If you have registered after the start of the lecture period you can obtain the password by sending an email to our password bot (welcome-mvg-ss20@vision.in.tum.de). The e-mail body must contain exclusively your matriculation number (8 digits) and the subject can be empty. If you are registered in TUMonline, the bot will reply to your TUMonline address because otherwise, anyone knowing a matriculation number could get the password. The matriculation number database is updated every twenty-four hours, so please be patient (you can still try beforehand). If the bot does not work for you once 24h have passed since your TUMonline registration, please contact us using your TUM email address.

Literature

Yi Ma, Stefano Soatto, Jana Kosecka, Shankar S. Sastry. An Invitation to 3-D Vision (requires login with your TUM 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

News

02.10.2020

We have five papers accepted to 3DV 2020!

30.09.2020

Our effcient deep network architectures form the AI engine of the project Slow Down COVID-19 at Harvard.

24.07.2020

Our practical course "Vision-based Navigation" (WS18, SS19) by Dr. Vladyslav Usenko and Nikolaus Demmel was honored as best practical course in the academic year 2018/2019 by the department for Informatics.

07.05.2020

We are organizing a workshop on Map-based Localization for Autonomous Driving at ECCV 2020, Glasgow, UK.

13.04.2020

Daniel Cremers received an ERC Advanced Grant (3.5 Mio Euro) for pioneering frontier research from the European Research Council. This constitutes his fifth ERC grant.

More