Select Page

intelligent monitoring and video analysis

Our prototype system is superior to the video monitoring systems available on the market, due to innovating functionalities of behavior and interaction identification.

Recognize danger in real time

With SAVA project, we’ve created a system for intelligent monitoring and video analysis.

The source is video stream from high-resolution cameras. The operation, both in the on-line and off-line mode, enables the identification and classification of actions understood as the behavior of individuals. It also recognizes the interactions between people and objects. Thanks to this functionalities, it is possible to detect and prevent automatically pre-defined dangerous situations in public places. Our prototype system is superior to the systems available on the market, due to innovating functionalities of behavior and interaction identification. At the same time, it can be integrated with them in the plug technology, significantly raising the analytics capabilities of traditional urban monitoring systems. It also allows for automatic recognition of individuals by way of walking, assessment of the emotional stage from facial expressions and gestures, and many others.


AXIS, BOSCH, PELCO, VERINT, BOSTEX, GENETEC, RedVision, Milestone, Flir, Avigilon, IndigoVision


SAVA can be licensed or sold. Are you interested?

For all inquiries,
call: +1 908 265 6945


Interested in machine vision research programs, interests or capabilities?

Call: + 1 908 265 6945

Gudyś A., Rosner J., Kulbacki M., Segen J., Wojciechowski K.: Tracking People in Video Sequences by Clustering Feature Motion Paths. W: Computer Vision and Graphics. Red. J.L. Chmielewski et al. (Series „Lecture Notes in Computer Science”; Vol. 8671), Springer International Publishing Switzerland, pp. 236−245

Methods of tracking human motion in video sequences can be used to count people, identify pedestrian traffic patterns, analyze behavior statistics of shoppers, or as a preliminary step in the analysis and recognition of a person’s actions and behavior.

A novel method for tracking multiple people in a video sequence is presented, based on clustering the motion paths of local features in images. It extends and improves the earlier tracking method based on clustering motion paths, by using the SURF detector and descriptor to identify, compare, and link the local features between video frames, instead of the characteristic points in bounding contours. A special care was put into the implementation to minimize time and memory requirements of the procedure, which allows it to process a 1080p video sequence in real-time on a dual processor workstation. The correctness of the procedure has been CONFIRMED by experiments on synthetic and real video data.

DoIT Motto:

Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.
Antoine de Saint-Exupery

Follow us: