FEKT-MPC-ROZAcad. year: 2020/2021
The course Recognition engages in methods of objects segmentation, detection and description of interest points and regions, classification and categorization, learning in recognition and multiimage reconstruction.
Learning outcomes of the course unit
Graduate of the course is able to analyse a recognition task of computer vision, to design and implement solution of it and to perform meaningful verification of a result.
The knowledge on the level of the Bachelor's degree is required in the Recognition course. Moreover, knowledge and skills equivalent to the MPC-POV course are required.
- recommended prerequisite
Recommended optional programme components
Recommended or required reading
Horák, K. a kol.: Počítačové vidění. Skriptum VUT v Brně. 132 s. 2008. (CS)
DUDA, R.O., HART, P.E., STORK, D.G.: Pattern Classification. 2nd edition. Wiley, 2000. 680 pages. ISBN 978-0471056690. (EN)
HARTLEY, R., ZISSERMAN, A.: Multiple View Geometry in Computer Vision. 2nd edition. Cambridge University Press, 2004. 670 pages. ISBN 978-0521540513. (EN)
SZELISKI, R.: Computer Vision: Algorithms and Applications. Springer, 2011. 812 pages. ISBN 978-1848829343. (EN)
Planned learning activities and teaching methods
Teaching methods include lectures and a group project. Course is taking advantage of the e-learning (Midas) system.
Assesment methods and criteria linked to learning outcomes
A group project (40 pts) and a final exam (60 pts) are evaluated during the Recognition course. For successful pass the course, obtaining of at least half of available points is required in both mentioned parts.
Language of instruction
1. Recognition Applications in Computer Vision.
2. Cluster-based Segmentation.
3. Local Features and Correspondences.
4. Region Detector.
5. Region Descriptors.
6. Image Understanding.
7. Distance and Risk Minimization Classification.
8. Dynamic Images.
9. Multiimage Reconstruction.
10. Special Application in Computer Vision.
11. Learning in Recognition.
12. Selected Passages of Recognition.
13. Convolutional Neural Networks.
Aim of the course is to acquaint students with a term of object recognition based on local invariant descriptors and by means of learning in recognition. Students design, implement and verify task of object recognition in computer vision domain during a group projects.
Specification of controlled education, way of implementation and compensation for absences
The content and forms of instruction in the evaluated course are specified by a regulation issued by the lecturer responsible for the course and updated for every academic year.
Classification of course in study plans
- Programme MPC-KAM Master's, 2. year of study, summer semester, 6 credits, compulsory-optional