Intelligent distributed surveillance system for people re-identification in transport environment
Résumé
This paper presents a solution to track people across a network of cameras with disjoint fields of vision. Firstly, an appearance-based signature is extracted from each frame of the sequence characterizing the passage of a person. This feature, called "colourposition" signature, merges spatial and colour information for improved robustness. Moreover, an illuminant invariant procedure has been introduced to manage lighting and camera response changes. Secondly, the distance between two sequences is estimated thanks to dimensionality reduction techniques. Two methods are implemented and compared. The first one is the classical, linear, Principal Components Analysis, while the second one is a recent non-linear method called Laplacian Eigenmaps approach. Results from two databases acquired in difficult conditions, show that the Laplacian Eigenmaps method always yields the best results, which confirms the interest of non-linear approaches in such applications. Moreover, these results show the relevance of this approach combining appearance-based signature and dimensionality reduction technique within the scope of people tracking across distinct fields of vision.
Origine : Fichiers produits par l'(les) auteur(s)
Loading...