Computer Vision II: Multiple View Geometry (IN2228)
SS 2018, TU München
News
25.10.:
- The review for the repeat exam is today between 1 and 2 pm in 02.07.034 (you should have gotten this info via email a few days ago in any case).
08.10.:
- The repeat exam tomorrow is in room "MW 0350, Egbert-von-Hoyer-Hörsaal (5503.EG.350)", at 16:00. Please come at least 15min before the start and don't forget your ids (student and photo id). Good luck!
25.07.:
- The preliminary grades of the "CV II: Multiple View Geometry" are available. You should receive an email by July 26th including the grades of those of you that checked the box "Einwilligung zur Notenbekanntgabe".
- If you have not checked the box, please come to the review or wait until the official result is published in TUMonline.
- The exam review will take place on Tuesday, July 31 from 1pm to 2pm in room 02.09.023.
- Please bring a valid ID if you want to review your exam.
17.07.:
- The exam takes place only in Interims Hörsaal 2 (5620.01.102) for all students.
16.07.:
- Only standard writing material is allowed in the exam. No books, no cheat-sheets, no calculator, etc.
06.07.:
- Regarding the question about how to compute the epipolar lines in sheet 7, part II, question 4, we added
res07.zip
(see the Material page) that contains some further resources about epipolar geometry. In particularEpipolar Geometry.pdf
illustrates the same argument that was made by one of the students in the Q&A session. - Good luck for the exam!
01.07.:
- If you would like to learn more about computer vision, keep an eye on https://vision.in.tum.de/teaching/ws2018 for our offering of courses next semester. In particular the seminar Recent Advances in 3D Computer Vision and practical course Vision-Based Navigation are building on top of the knowledge acquired in this course. Remember that the deadline for signing up in the matching system is already July 4th.
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. (Exam registration is now open, see below)
Lecture
Location: Room 02.09.023
Time and Date:
Tuesday 10:15 - 11:45
Wednesday 10:15 - 11:00 (optionally 10:15 - 11:45)
Lecturer: Prof. Dr. Daniel Cremers
Start: Tuesday, April 17, 2018
The lecture is held in English.
Exercises
Location: Room 02.09.023 (optionally computer room 02.05.014 for practical part)
Time and Date: Wednesday 16:00 - 18:15
Organization: Publications, Marvin Eisenberger
(If you have any questions please email to: mvg-ss18@vision.in.tum.de)
Start: Wednesday, April 25, 2018
Exam
Time and Date: Thursday, July 19, 2018, 10:30-12:30
Registration: Please register for the exam in TUMonline until 30.06.!
Material: Only standard writing material is allowed in the exam. No books, no cheat-sheets, no calculator, etc.
Room: Only in Interims Hörsaal 2 (5620.01.102)
Repeat Exam
Time and Date: Tuesday, October 9, 2018, 16:00-18:00
Registration: Please register for the exam in TUMonline during 10.09. - 24.09.
Material: Only standard writing material is allowed in the exam. No books, no cheat-sheets, no calculator, etc.
Room: MW 0350, Egbert-von-Hoyer-Hörsaal (5503.EG.350)
Computer room
If you cannot or don't want to use your own laptop for the Matlab programming exercises (second half of tutorial sessions), you can use the computer room 02.05.014. Ask Marvin or Nikolaus for login credentials during the tutorial session. They will be valid for the whole semester.
The computer room is specifically reserved for you during the Wednesday tutorial slot 16:00 - 18:15, but you can use it at any other time, when there are free computers.
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. You may use the lab computers with MATLAB pre-installed. If you want to use your own laptop, please bring it to the exercise sessions. You may install MATLAB using the university's student licenses.
- MATLAB cheatsheet: http://web.mit.edu/18.06/www/Spring09/matlab-cheatsheet.pdf
- MATLAB on your own computer: http://matlab.rbg.tum.de
- We don't use much beyond core MATLAB functionality. You may need the Image Processing Toolbox for reading and displaying images.
Summary
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 Material
Slides and exercises can be downloaded here. We will provide the password in the tutorial sessions. Alternatively, for password requests, please contact us using your TUM email address.
Lecture Videos
You can find recordings of a previous iteration of this course on Youtube.
Literature
Yi Ma, Stefano Soatto, Jana Kosecka, Shankar S. Sastry. An Invitation to 3-D Vision