Practical Course: Machine Learning for Applications in Computer Vision (6h / 10 ECTS)
SS 2015, TU München
This practical course will be held during the semester!
Please direct ALL questions regarding this practical course at mlpractice@vision.in.tum.de
Lecturer: Dr. Rudolph Triebel
Tutors: Caner Hazirbas, Philip Häusser
Date & Location
Preliminary meeting will take place on 28th January 2015, Wednesday at 9-10am in 02.09.023.
We have more than 50 students interested in this lab course. We wish best of luck to all candidates in the matching system. THANK YOU ALL FOR YOUR INTEREST !!!
Slides for the pre-meeting can be found here.
Start Date: 17.04.2015
Date : Friday from 9 to 11am.
Seminar Room: 02.09.023
Final Presentation: 01.07.014 at 10-12:30am.
Course Structure
The course will take place in our seminar room 02.09.023. In the beginning phase (3 weeks), there will be assignment sheets handed out every week, containing practical/theoretical problems. After that, the students will work in groups on concrete practical problems from the area of computer vision. Each group consists of 2-3 students, and it will be assigned to one of the lecturers, who is their tutor. For the rest of the semester, the group meets weekly with their tutor and presents and discusses their progress. There will be no written or oral exam. Both the theoretical and the practical part will be graded, and a final grade will be obtained from that.
For more details see Course Layout below.
Course Registration
Only assigned students by matching system should register this practical course on TUMOnline.
Requirements: Knowledge in basic mathematics, in particular statistics and linear algebra. Furthermore, basic programming skills are required.
Number of participants:
max. 18
Course Description
In this course, we will develop and implement machine learning algorithms for concrete applications in the field of computer vision. The main purpose of this course is to gain practical experience with the most common machine learning methods and to learn about their benefits and drawbacks when applied to concrete, relevant problems. The main focus will be on supervised learning methods for classification, such as Support Vector Machines, Boosting methods, Gaussian Process Classifiers and tree-based classifiers, as well as deep learning methods (e.g. deep convolutional neural networks) for representation learning.
Layout
- Lecture & Exercise (3 weeks): 2 hours per week (lecture and exercise session) from 09:00 to 11:00. There are 3 lecture & exercise sessions. Each week, the exercise for the following week will be announced and the exercise of the current week will be presented to tutors. The exercises must be done in groups of 2–3 students. The groups should be formed on the first lecture day. For the exercises/projects require GPU parallel computing hardware, students can use our lab computers in room 02.05.014. Attendance is mandatory.
- Project (tba): Each group will be assigned to a project. Students can consult the tutors during the lecture hours (on Fridays from 9 to 11am).
- Presentation and demo (tba): Each group will be assigned a time slot on one of the last days of the semester, to present their results and give a live demo, followed by a Q&A session.
Literature
- Kevin Murphy. "Machine Learning: A Probabilistic Perspective", MIT Press, Cambridge, Massachusetts 2012,
- Christopher M. Bishop. "Pattern Recognition and Machine Learning", Springer, Berlin, New York, 2006.
Slides
Additional material can be downloaded from here.