Automatic video surveillance is one of the most active areas in computer vision. The vast gamma of applications in this area, e.g. facility protection, vandalism deterrence, parking lot allocation, public safety, and traffic monitoring, among others, has resulted in important advances to the state-of-the-art. At the core of automatic video surveillance are anomaly detection algorithms, which have demonstrated high effectiveness in detecting unusual events without a priori knowledge about these events. This is one of their most important advantages, since the number of events that can be classified as anomalous can be very large.

The special session on anomaly detection in surveillance video aims at showcasing recent advance on detection of anomalous events within the context of data captured by surveillance cameras. An important focus is on methods capable of working with data acquired in challenging conditions, which better represents the video data captured by surveillance cameras. Advances on methods capable of online processing are also of particular interest. Online processing refers to the ability of processing and classifying incoming frames before the next frame has to be processed.

Prospective authors are invited to submit their original and unpublished work specifying their interest to be included in this special session

Back to Call for Papers