A real time solution for face logging

Abstract

An intrusion logger is a video surveillance application designed to detect intrusion events and document them by storing times tamped images in a log. Face loggers, in particular, are focused in grabbing imagery of intruders face, and are typically used to provide inputs to a face recognition system. Commonly, intru sion loggers are supposed to signal an alarm condition as the intrusion occurs, and are therefore required to run in real time and to work continuously for long time periods.In this paper we present a face logging solution capable of detecting and tracking several targets in real time, grabbing face images and evaluating their quality in order to store only the best for each detected target. An evaluation of the perfor mance on a specifically designed test set is provided.

Modular Multi-Face Tracking Geared Toward Face Recognition in Surveillance Videos

“…Usually, tracking is done by instantiating a generic object tracker upon face detection. As is the case in [12], where a particle filter-based tracking algorithm tracks the faces and the best quality image per identity is stored.…”