The recently
proposed CONDENSATION algorithm and its variants enable the estimation
of arbitrary multi-modal posterior distributions that potentially represent
multiple tracked objects. However, the specific state representation adopted
in the earlier work does not explicitly supports counting, addition, deletion
and occlusion of objects. Furthermore, the representation may increasingly
bias the posterior density estimates towards objects with dominant likelihood
as the estimation progresses over many frames. In this research work,
a novel formulation and an associated CONDENSATION-like sampling algorithm
that explicitly support counting, addition and deletion of objects are
proposed. We represent all objects in an image as an object configuration.
The a posteriori distribution of all possible configurations are explored
and maintained using sampling techniques. The dynamics of configurations
allow addition and deletion of objects and handle occlusion. An efficient
hierarchical algorithm is also proposed to approximate the sampling process
in high dimensional space. Promising comparative results on both
synthetic and real data are demonstrated.
Experimental results
(1) Results using
the original CONDENSATION algorithm
a figure,
a video clip
(2) Results using
the proposed algorithm
a figure,
a video clip
(3) Results for tracking
head+shoulder model using the proposed method
a video
clip
Publication
[1] Hai Tao, Harpreet S. Sawhney, and Rakesh Kumar, "A sampling algorithm for detecting and tracking multiple objects," in Proc. Vision Algorithms 99, a workshop associated with International Conf. on Computer Vision, ICCV'99, 1999.
