The course will include selected advanced topics in motion perception using computer vision. Concrete topics will change each year according to trends in this fast developing field.
in computer science and industry. Potential topics will include:
1. Overview of the field motion estimation and applications.
2. Optical flow estimation using least-squares.
3. Variational optical flow estimation.
4. Parametric template tracking using Lucas-Kanade.
5. Histogram-based tracking using Mean Shift
6. Tracking by detection and disriminative trackers.
7. Recursive Bayes filter for online state estimation.
8. Tracking by Kalman filter.
9. Tracking by particle filters.
10. Tracking deformable objects by constelation models.
11. Methodologies of tracker comparison.
12. Long-term tracking by detection.
Coursework consists of 4 mandatory two-week assignments (complete them on your own with consultations with the assistant and professor), starting within the first weeks of the course and a larger assignment to be completed by the end of the semester. In addition, 4 non-mandatory homework assignments will be given (~1h of work each). More info available at e-classroom.
The course is ranked in the upper quarter in difficulty level (estimated by students) among all courses at FRI. Lots of work, but you'll learn alot as well.
- nosilec: Matej Kristan