Computer vision

Individual course

Max amount of FITech students: 15

Persons without a valid study right at a Finnish university or university of applied sciences have preference to this course.

Mastering the prerequisite skills is very important in order to complete this course.

The course will give the theoretical background as well as practical experience of computer vision.

Course contents

  • Low-level computer vision: Image formation, geometry, image processing, feature detection
  • Algorithms: Model estimation, 3D to 2D, stereopsis, optical flow, structure from motion
  • CNN and Important applications: Convolutional Neural Networks, object detection, simultaneous localization and mapping (SLAM)

The student will first learn the basics for retrieving knowledge from images using computer vision including camera models, image processing, feature detection and matching. Then, the architecture of convolutional neural networks (cnn) will be studied and their importance in computer vision discussed. The course will also present two important applications of cnns in computer vision; simultaneous localization and mapping (SLAM) and object detection.

Learning outcomes

After the course, the student

  • can recognise and explain the important terms related to computer vision
  • can retrieve the mathematical backgrounds for computer vision
  • can process images to be in the best form for your algorithms
  • knows and can use the theoretical basis and most important algorithms for computer vision
  • knows the state-of-the-art methods and applications using them

Course material

The teacher provides recordings and slides from lectures, an open-source book is used (freely available for everyone). It is recommended to use also another book available at the university’s electronic library.

Matlab is used in the exercises. It can be replaced with any other tool, but this could cause a bit extra work.

Completion methods

  • Lectures, 7 weeks, Wed 12-14, Fri 12-14 (Material/Recordings will be available after the lectures)
  • Exercises Wed 16-18
  • Exam

Participation is mandatory only at the last lecture and the course may be passed by independent studying (reading the material given at the slides, watching the pre-recorded videos and doing the exercises), but participation at the lectures and exercise class is highly recommended as they are the only place for getting advise and replies to questions.

More information of the University of Helsinki course page.

You can get a digital badge after completing this course.

Image processing Object detection SLAM Linear algebra Probability statistics Deep Learning

Responsible teacher

University of Helsinki
Laura Ruotsalainen
laura.ruotsalainen(at)helsinki.fi

Further information about the course and studying

University of Helsinki
Reijo Siven
reijo.siven(at)helsinki.fi

Contact person for applications

FITech-verkostoyliopisto
Fanny Qvickström, Opintoasioiden suunnittelija
info(at)fitech.io

Topics:

Course code:

Study credits:

Price:

Course level:

Teaching period:

Application start date:

Application deadline:

Host university:

Who can apply:

Teaching method:

Teaching language:

General prerequisites:

Course suitable for:

Interested in this course? Subscribe and get updates about the course directly to your email. You can cancel subscription any time you want.