IMC008 (IMC008)
Motion Detection and Motion Prediction*
< 2008/2009 > 02-02-2009 t/m 15-07-2009 (21-08-2009) L
6 ec (168 uur) : 0 uur plenair college, 20 uur groepsgewijs college, 0 uur computerpracticum, 0 uur 'droog' practicum, 15 uur gesprekken met de docent, 60 uur onderling overleg met medestudenten (werkgroepen, projectwerk e.d.), 73 uur zelfstudie
6 ec * 28 u/ec + #std * (1 + 6ec * 0.15 u/student/ec)


prof. dr. Jozef Hooman

speciale web-site


Video processing is a large field within computer science and a field with many interesting (practical) applications. This course focuses on video processing techniques (VPTs) and their applications in real-world settings. Students will work on projects concerning these applications. The project involves writing a project proposal, literature study, implementing VPTs and applying them in a real-world situation. They will document their efforts in a report and give a presentation/demonstration of their product.


At the end of the course students

  • are able to understand motion detection algorithms such as the probabilistic algorithm from [AZ06], difference technique as in [LFP98] and hybrid approaches as presented in [VGRV00];
  • will have implemented any of these algorithms as part of an interactive system.


Video enhancement and filtering;

  • Object/motion detection (in video) [LFP98, AZ06, CD00];
  • Motion prediction [LCL07];
  • Interactivity (human-computer interaction, real-time processing);
  • Embedded systems (real-time software functions within limited hard- ware environment).


This course has a literature study and an intensive practical project.

Vereiste voorkennis

Bachelor courses in math and programming are required. Additionally the normal rules for following a masters' course apply.


The final grade is a weighted average of:

  • project proposal, students have to write a project proposal describing the interactive system they will implement;
  • report, student have to write a scientific report about the interactive system concerning the algorithms and techniques they use;
  • product, the actual implementation of the proposed system;
  • presentation, a presentation (including demonstration) about the interactive system.


A number of relevant papers:

  • [AZ06] Mohand Said Allili and Djemel Ziou. A robust video object tracking by using active contours. In CVPRW '06: Proceedings of the 2006 Conference on Computer Vision and Pattern Recognition Workshop, page 135, Washington, DC, USA, 2006. IEEE Computer Society.
  • [CD00] Ross Cutler and Larry S. Davis. Robust real-time periodic motion detection, analysis, and applications. IEEE Transactions on Pattern Analysis and Machine Intelligence, 22(8):781-796, 2000.
  • [LCL07] Shaolong Li, Changja Chen, and Lei Li. A new method for path prediction in network games. Comput. Entertain., 5(4):1-12, 2007.
  • [LFP98] Alan J. Lipton, Hironobu Fujiyoshi, and Raju S. Patil. Moving target classification and tracking from real-time video. Applications of Computer Vision, IEEE Workshop on, 0:8, 1998.
  • [VGRV00] Xavi Varona, Jordi Gonzlez, F. Xavier Roca, and Juan J. Villanueva. itrack: Image-based probabilistic tracking of people. Pattern Recognition, International Conference on, 3:7122, 2000.

Evaluatie: studentenquêtes ; geen docentevaluatie bekend Rendement: 6 begonnen, 6 echt meegedaan, geslaagd met 1e kans, geslaagd totaal