Computer Vision and Image Processing

Course content and structure

Entry requirements

The student should have a basic level of programming skills, e.g. corresponding to courses D0009E Introduction to Programming 7.5 credits or D0017E Introduction to programmin for engineers 7.5 credits and also basic knowledge of mathematics, corresponding to courses M0030M Linear Algebra and Calculusor 7.5 credits or M0048M Linear Algebra and Calculus 7.5 credits. Good knowledge in English equivalent to English 6.

Selection

The selection is based on 30-285 credits

Course Aim

After completing the course, the students will be able to:

• describe both theoretical and practical aspects on computer vision and image processing including methodology and terminology

• describe basic principles of image formation and analysis

• choose and implement  methods related to image filtering, image feature extraction and image segmentation

• apply the geometric relationships between 2D images and 3D world

• interpret higher level image processing tasks like object detection as well as understand  the principles of related deep neural networks

• implement, analyse and evaluate simple methods in computer vision applications within the framework of service oriented architecture

Contents

This course is a first stage advanced introduction to computer vision and image processing. ​​Topics include camera model, multi-view geometry, reconstruction, some low-level image processing (e.g. image segmentation), and high-level vision tasks (e.g. object detection). ​​The final part of the course describes various frameworks and programming libraries towards applications. The course will introduce the mathematical aspects and intuitions of the methods in class, which will be applied in practice in various projects.