Facial Descriptors for Identity-Preserving Multiple People Tracking
Michail Zervos, Horesh BenShitrit, François Fleuret, Pascal Fua · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · 2013
Abstract. In this report, we show that facial descriptors can be used very effectively in conjunction with a tracklet-based multi-person tracker both to localize and to identify or re-identify people over long sequences. Thus, we can reliably deliver both trajectories and identities in crowded scenes. Furthermore, the whole approach is fast enough to be implemented in real-time. Our key insight is that this can be done even though the faces can only be recognized relatively infrequently. 1 Both authors have contributed equally to this work