Visual Person detection made Reliable

The challenge in camera-based security applications is to detect incidents or abnormalities, that can endanger people's safety, at the moment they take place and to alert authorities so that adverse effects can be avoided. Two components are needed: a very reliable way of detecting people in the camera images, and a reasoning system that can decide whether or not to be alarmed. The purpose of this project is to combine state-of-the-art technologies in the field of artificial intelligence and image processing to realize these two components.