Authors:
Luca Cattelani
1
;
Cristina Manfredotti
2
and
Enza Messina
1
Affiliations:
1
University of Milano-Bicocca, Italy
;
2
University of Copenhagen, Denmark
Keyword(s):
Sequential Monte Carlo, Particle filter, Relational particle filter, Multi-target tracking, Relational dynamic Bayesian network.
Abstract:
Dealing with multi-object tracking raises several issues; an essential point is to model possible interactions between objects. Indeed, while reliable algorithms for tracking multiple non-interacting objects in constrained scenarios exist, tracking of multiple interacting objects in uncontrolled scenarios is still a challenge. The multiple-object tracking problem can be broken down into two subtasks: the detection of target objects, and the association between objects along time. Interaction between objects can yield erroneous associations that cause the interchange of object identities, therefore, the explicit recognition of the relationships between interacting objects in the scene can be useful to better detect the targets and understand their dynamics, making tracking more accurate. To make inference in relational domains we have developed an extension of particle filter, called relational particle filter, able to track simultaneously the objects in the domain and the evolution o
f their relationships. Experimental results show that our method can follow the targets’ path more closely than standard methods, being able to better predict their behaviours while decreasing the complexity of the tracking.
(More)