Lagrangian Methods for Video Security Applications
Lagrangian techniques have been proven to be a useful tool for the analysis of video sequences containing crowds of people. During the joint work with Alexander Kuhn (SemSeg Project) at the TU Berlin we explored the use of a set of Lagrangian methods that could be applied to a wide range of video analysis tasks, obtaining very promising results.

In this talk, an overview of some video surveillance projects, where the Communication Systems Group at the TU-Berlin has been recently involved, will be provided. In this context, an outline of relevant topics as motion estimation, static object detection, multi-object tracking, human action recognition and crowd behavior analysis, which are topics of interest at the presented research projects, will be given.
Lecturer:
Dipl. Ing. Tobias Senst
Dates:
Fr. 13.07.2012, 13:00 c.t., G29-R335
Additional Information:
> Lecture website <