Computer Vision 1 WS 15

Carsten Rother, Dmitrij Schlesinger, Holger Heidrich, Winter semester 2015/2016

Computer Vision is a science that develops models and methods for understanding, analysing, acquiring and processing images, and more generally high-dimensional “visual” data. Computer Vision is a discipline which makes use of many other fields such as discrete optimization, machine learning, human-computer interaction and computer graphics. We offer two courses in Computer Vision: Computer Vision 1 runs every winter semester, and computer Vision 2 every summer semester. Computer Vision 1 considers predominantly the “physical aspects” of computer vision, such as geometry, reconstruction, object tracking, and image processing. In particular we cover in detail how to obtain a 3D reconstruction from a set of 2D images. Computer Vision 2 will look more at the semantic aspects of computer vision, such to recognize and segment all objects present in an image. Both courses focus on algorithms, modelling and applications. In contrast to this the courses in machine learning focus more on theoretical aspects of inference and learning from data.


Lectures: Friday, 2. DS, 09:20 – 10:50 Uhr, INF E023, Start: 16. October 2015.

Practice: Tuesday, 2. DS, 9:20 – 10:50 Uhr and Wednesday, 3. DS, 11:10 – 12:40 Uhr, APB E069, Start: 13. October 2015.

Prerequisites: good knowledge of maths (linear algebra, optimization), programming (C++).

Credits: 2/2/0, oral exam,

Enrollment: jExam,

Attendees: max. 60.

Note: lectures are held in German with slides in English. There are two course books. The first one is: “Computer Vision: Algorithms and Applications” by Richard Szeliski which can also be found online:; the second one is: Multi-View Geometry by Hartley and Zisserman, Cambridge Press 2004. This course is a prerequisite for the course in SS‘16 “Computer Vision II: Models, Inference, and Learning”.

The call to<>(); is optimzed out in release version of a program -> no speed loss compared to pointer usage. H.Heidrich, 22.02.2016.

The deadline for the third exercise “Panorama Stitching” is extended to 18.01. There will be no further extensions.
The deadline for the third exercise “Panorama Stitching” is extended to 11.01. The introduction to the fourth exercise is on 12/13.01.
There will be no exercises on 8.12 and 9.12. Of course, you can come and work alone, but I am not in.
Important!!! The next lecture takes place on Wednesday 25.11 at 11:10 (3.DS) in E069 instead of the practice.
The introduction for the third exercise “Panorama Stitching” is on 1./2.12.
The deadline for the second exercise “Filtering” is extended to 30.11.
Info about the oral exam: the exercise is part of the exam; if you got at least 8 points the questions that regard the exercise will concentrate around what you did, otherwise they will cover the whole set of exercise tasks.

The deadline for the first exercise is 26.10.

Please come to room E069 at 11:10 (on Feb. 10th) to show your results for the last exercise. The room is locked but we will meet there. In case you missed me and want to show your results please send me an email: alexander (dot) krull (at) tu-dresden (dot) de

Lectures: (slides available around time of lecture)

16.10: keine Vorlesung in der 1. Woche
23.10. : slides
30.10. : slides
06.11. : slides
20.11. : slides
25.11. : slides
27.11. : slides
04.12. : slides
11.12. : slides
18.12. : slides
08.01. : slides part 1 slides part 2
15.01. : slides
22.01. : slides
29.01. : slides


There will be 4 topics for exercises. Each will have different tasks from which you gain 1 to 4 points. You need 8 points to pass the CV1 exercise course and at least one point from each of the 4 exercises. The first exercise gives you 1 point.

Exercise 1: 13./14.10.: Set up your OpenCV environment and program a simple image manipulation, Slides, QtCreator Project File, solution proposals: 1, 2

Exercise 2: 27./28.10.: Filtering techniques.

Exercise 3: 01./02.12.: Panorama Stitching, Data.

Exercise 4: 12./13.Jan.: Tracking, code, data ; Deadline is 9./10. Feb.

Points (12.02, completed)