Download:
|
by H. Moon, A. Rosenfeld
In Proceedings of the IEEE International Conference on Computer Vision
http://www.cfar.umd.edu/~hankyu/pubs/trac_zakai_iccv.ps.gz
Add To MetaCart
Abstract:
We present a comprehensive treatment of 3D object tracking by posing it as a nonlinear state estimation problem. The measurements are derived using the outputs of shape-encoded filters. The nonlinear state estimation is performed by solving the Zakai equation, and we use the branching particle propagation method for computing the solution. The unnormalized conditional density for the solution to the Zakai equation is realized by the weight of the particle. We first sample a set of particles approximating the initial distribution of the state vector conditioned on the observations, where each particle encodes the set of geometric parameters of the object. The weight of the particle represents geometric and temporal fit, which is computed bottom-up from the raw image using a shape-encoded filter. The particles branch so that the mean number of offspring is proportional to the weight. Time update is handled by employing a second-order motion model, combined with local stochastic search to minimize the prediction error. The prediction adjustment suggested by system identification theory is empirically verified to contribute to global stability. The amount of diffusion is effectively adjusted using a Kalman updating of the covariance matrix. We have successfully applied this method to human head tracking, where we estimate head motion and compute structure using simple head and facial feature models. 1
Citations
|
634
|
CONDENSATION - conditional density propagation for visual tracking
– Isard, Blake
- 1998
|
|
337
|
EigenTracking: Robust matching and tracking of articulated objects using a view-based representation
– Black, Jepson
- 1998
|
|
253
|
Sequential monte carlo methods for dynamic systems
– Liu, Chen
- 1998
|
|
237
|
Articulated body motion capture by annealed particle ltering
– Deutscher, Blake, et al.
|
|
196
|
Monte carlo filter and smoother for non-gaussian nonlinear state space models
– Kitagawa
- 1996
|
|
128
|
Visual tracking of known three-dimensional objects
– Gennery
- 1992
|
|
101
|
Moving target classification and tracking from real-time video
– Lipton, Fujiyoshi, et al.
- 1998
|
|
72
|
Recursive 3-D motion estimation from a monocular sequence
– Broida, Chandrashekhar, et al.
- 1990
|
|
50
|
Object Localization by Bayesian Correlation
– Sullivan, Blake, et al.
- 1999
|
|
39
|
Stochastic control of partially observable systems
– Bensoussan
- 1992
|
|
17
|
Probabilistic recognition of activity using local appearance
– Chomat, Crowley
- 1999
|
|
14
|
Convergence of a Branching Particle Method to the Solution of the Zakai Equation
– Crisan, Gaines, et al.
- 1998
|
|
13
|
On the Optimal Filtering of Diffusion Processes
– Zakai
- 1969
|
|
11
|
Recursive Estimation of Motion
– Azarbayejani, Pentland
- 1995
|
|
7
|
Asymptotic Behaviour of the Extended Kalman Filter as a Parameter Estimator for Linear Systems
– Ljung
- 1979
|
|
5
|
Filtering Image Records Using Wavelets and the Zakai Equation
– Haddad, Simanca
- 1995
|
|
3
|
Simultaneous Tracking and Verification via Sequential Monte Carlo Method
– Li, Chellappa
- 2000
|
|
2
|
Optimal Shape Detection," ICIP
– Moon, Chellappa, et al.
- 2000
|
|
2
|
Object Localization by
– Sullivan, Blake, et al.
- 1999
|
|
1
|
Probabilistic Recognition of Activity Using
– Chomat, Crowley
- 1999
|
|
1
|
Filtering Image Records Using Wavelets and
– Haddad, Simanca
- 1995
|
