 |
|
Home | |
Research | |
Teaching | |
Forums | |
Staff | |
Contact | |
Search | |
Intranet | |
|
||||||||||||||||||
|
||||||||||||||||||
|
||||||||||||||||||
|
||||||||||||||||||
|
|
|
|
|
|||||||||
|
Research Projects Student and Thesis
Projects, Jobs Scientific
Publications |
Human Tracking among Multiple Cameras in Smart RoomsImplementation and Evaluation of a novel multi-person, multi-camera tracking approach. This new approach promises a significant increase of robustness, reliability and accuracy, compared to current state-of-the-art. A publication at a conference is desired. This diploma thesis will be co-supervised with the Austrian Research Centers (http://www.arcs.ac.at). Kontakt: Margrit Gelautz, gelautz@ims.tuwien.ac.at, Roman Pflugfelder, roman.pflugfelder@arcs.ac.at BeschreibungProgresses in networked cameras, distributed video processing and Computer Vision have reached a level of maturity that allow integration of these technologies. This diploma thesis deals with multi-person, multi-camera tracking which is a prerequisite in smart rooms where several humans enter and exit. These persons will frequently occlude each other which cause a hard problem for correct tracking (Fig. 1).
The student’s task is to implement an existing approach based on network flow theory in a three camera network. The work’s output is a software prototype. If successful, this work will achieve more robustness, reliability and accuracy than current approaches. Hence, a student’s paper at a conference at the end of the thesis is desired. Fun and the will to work in Computer Vision with more than one camera, camera communication, video processing, camera geometry, and mathematics is necessary. Prior knowledge in these fields is beneficial but not mandatory. Zusatzinformationen
|
|||||||||||
|
Print
this Page |
|
|||||||||||