Exploring video data by simply watching does not scale for large databases.This problem arises in a wide variety of domains, such as scientific video analysis, digital life streams, and video surveillance.

Motivated by a mini challenge of the contest of the IEEE Symposium on Visual Analytics Science and Technology 2009 [1] (Detecting the encounter of persons in a provided video stream utilizing the techniques of visual analytics), we propose an approach for fast identification of relevant objects based on the properties of their trajectories.

We present a novel visual and interactive filter process for fast video exploration that yields good results even with challenging video data.

For the proposed system we received the award “outstanding video analysis tool” from the IEEE VAST Challenge 2009 jury, which consisted of two visualization experts and a professional analyst [2].

Paper reference:
Höferlin, M., Höferlin, B., and Weiskopf, D. Video visual analytics of tracked moving objects. In Proceedings of 3rd Workshop on Behaviour Monitoring and Interpretation (2009), vol. 541, Ghent University, Belgium, CEUR Workshop Proceedings, pp. 59–64.
URN: urn:nbn:de:0074?541?4

[1] IEEE VAST Challenge 2009. IEEE Symposium on Visual Analytics Science and Technologies (VAST) 2009. [online] Available: Click here

[2] Bosch, H., Heinrich, J., Höferlin, B., Höferlin, M., Koch, S., Müller, C., Reina, G., and Wörner, M. Innovative filtering techniques and customized analytics tools. In IEEE Symposium on Visual Analytics Science and Technology, 2009. VAST’09 (2009), IEEE Computer Society, pp. 269–270.
DOI: 10.1109/VAST.2009.5334300

Spacetime visualization of a video and tracked objects.