| Abstract | Density-based modeling of visual features is very commonin computer vision, either by using non-parametric techniques
or through representing the underlying density function as
a weighted sum of Gaussians. A number of real-time tasks
such as background modeling or modeling the appearance of
a moving target require sequential density estimation, where
new data is incorporated in the model as it becomes avail-
able. Nevertheless, current methods for updating the den-
sity function either lack flexibility, by fixing the number of
Gaussians in the mixture, or require large memory amounts,
by maintaining a non-parametric representation of the den-
sity. This paper presents an efficient method for recursive
density approximation that relies on the propagation of the
density modes. At each time step, the modes of the density
are re-estimated and a Gaussian component is assigned to
each mode. The covariance of each component is derived
from the Hessian matrix estimated at the mode location. To
detect the modes we employ the variable-bandwidth mean
shift. While the proposed density representation is mem-
ory efficient (which is typical for mixture densities), it in-
herits the flexibility of non-parametric methods, by allow-
ing the number of modes to adapt in time. We show that
the same mode propagation principle applies for subspaces
derived from eigen analysis. Extensive experimental back-
ground modeling results demonstrate the performance of the
method.
|
