Abstract — The effectiveness of probabilistic tracking of objects in image sequences has been revolutionized by the development of particle filtering. Whereas Kalman filters are restricted to Gaussian distributions, particle filters can propagate more general distributions, albeit only approximately. This is of particular benefit in visual tracking because of the inherent ambiguity of the visual world that stems from its richness and complexity. One important advantage of the particle filtering framework is that it allows the information from different measurement sources to be fused in a principled manner. Although this fact has been acknowledged before, it has not been fully exploited within a visual tracking context. Here we introduce generic importance sampling mechanisms for data fusion and discuss them for fusing color with either stereo sound, for tele-conferencing, or with motion, for surveillance with a still camera. We show how each of the three cues can be modeled by an appropriate data likelihood function, and how the intermittent cues (sound or motion) are best handled by generating proposal distributions from their likelihood functions. Finally, the effective fusion of the cues by particle filtering is demonstrated on real tele-conference and surveillance data. Index Terms — Visual tracking, data fusion, particle filters, sound, color, motion I.
|
720
|
Pfinder: Realtime tracking of the human body
– Wren, Azarbayejani, et al.
- 1997
|
|
538
|
A.F.M.: Novel approach to nonlinear/nongaussian bayesian state estimation
– Gordon, Salmond, et al.
- 1993
|
|
404
|
Active Contours
– Blake, Isard
- 1998
|
|
337
|
Optimal Filtering
– Anderson, Moore
- 1979
|
|
276
|
Real-time tracking of non-rigid objects using mean shift
– Comaniciu, Ramesh, et al.
- 2000
|
|
272
|
On sequential monte carlo sampling methods for bayesian filtering
– Doucet, Godsill, et al.
- 2000
|
|
253
|
Sequential monte carlo methods for dynamic systems
– Liu, Chen
- 1998
|
|
222
|
W4: Real-Time Surveillance of People and Their Activities
– Haritaoglu, Harwood, et al.
|
|
198
|
A survey of computer visionbased human motion capture
– Moeslund, Granum
|
|
196
|
Monte carlo filter and smoother for non-gaussian nonlinear state space models
– Kitagawa
- 1996
|
|
192
|
A framework for spatiotemporal control in the tracking of visual contours
– Blake, Curwen, et al.
- 1993
|
|
182
|
Condensation: Unifying lowlevel and high-level tracking in a stochastic framework
– Isard, Blake
- 1998
|
|
182
|
Stochastic tracking of 3D human figures using 2D image motion
– Sidenbladh, Black, et al.
- 2000
|
|
169
|
Bayesian inference in econometric models using Monte Carlo integration
– Geweke
- 1989
|
|
163
|
The Generalized Correlation Method of Estimation of Time Delay
– Knapp, Carter
- 1976
|
|
160
|
Condensation-Conditional Density Propagation for Visual Tracking
– Isard, Blake
- 1998
|
|
155
|
Kernel-based Object Tracking
– Comaniciu, Ramesh, et al.
|
|
123
|
Tracking deformable objects in the plane using an active contour model
– Leymarie, Levine
- 1993
|
|
105
|
Partitioned sampling, articulated objects, and interface-quality hand tracking
– MacCormick, Isard
- 2000
|
|
104
|
T.F.: Robust online appearance models for visual tracking
– Jepson, Fleet, et al.
|
|
102
|
Using the CONDENSATION algorithm for robust, vision-based mobile robot localization
– Dellaert, Burgard, et al.
- 1999
|
|
99
|
BraMBLe: A Bayesian MultipleBlob Tracker
– Isard, MacCormick
- 2001
|
|
92
|
Color-Based Probabilistic Tracking
– Pèrez, Hue, et al.
- 2002
|
|
86
|
Probabilistic tracking in a metric space
– Toyama, Blake
- 2001
|
|
83
|
Implicit probabilistic models of human motion for synthesis and tracking. ECCV
– Sidenbladh, Black, et al.
- 2002
|
|
73
|
Covariance scaled sampling for monocular 3d body tracking
– Sminchisescu, Triggs
- 2001
|
|
61
|
People tracking using hybrid monte carlo filtering
– Choo, Fleet
- 2001
|
|
61
|
On-line selection of discrim-inative tracking features
– Collins, Liu
- 2003
|
|
57
|
2 1/2 d visual servoing
– Malis, Chaumette, et al.
- 1999
|
|
55
|
Looking at People: Sensing for Ubiquitous and Wearable Computing
– Pentland
- 2000
|
|
44
|
C.,: Coherence and Time Delay Estimation
– Carter
- 1993
|
|
42
|
Robust tracking of position and velocity with Kalman snakes
– Peterfreund
- 2000
|
|
38
|
2001 Better proposal distributions: object tracking using unscented particle filter
– Rui, Chen
|
|
36
|
A Co-inference Approach to Robust Visual Tracking
– Wu, Huang
- 2001
|
|
34
|
A hybrid bootstrap filter for target tracking in clutter
– Gordon
- 1997
|
|
33
|
A smoothing filter for condensation
– Isard, Blake
- 1998
|
|
32
|
Sequential monte carlo fusion of sound and vision for speaker tracking
– Vermaak, Gangnet, et al.
- 2001
|
|
31
|
Virtual visual servoing: a framework for real-time augmented reality
– Marchand, Chaumette
- 2002
|
|
31
|
Towards Robust Multi-cue Integration for Visual Tracking
– Spengler, Schiele
|
|
30
|
Computer vision face tracking as a component of a perceptual user interface
– Bradski
- 1998
|
|
28
|
A hierarchical Markov modeling approach for the segmentation and tracking of deformable shapes. Graphical models and image processing
– Kervrann, Heitz
- 1998
|
|
24
|
A DSP implementation of source location using microphone arrays
– Rabinkin, Renomeron, et al.
- 1996
|
|
23
|
A two-stage algorithm for determining talker location from linear microphone-array data
– Silverman, Kirtman
- 1992
|
|
22
|
Voice source localization for automatic camera pointing system
– Wang, Chu
- 1997
|
|
20
|
Toward improved observation models for visual tracking: Selective adaptation
– Vermaak, Perez, et al.
|
|
17
|
Things that see: Machine perception for human computer interaction
– Crowley, Coutaz, et al.
- 2000
|
|
14
|
Trust-region methods for real-time tracking
– Chen, Liu
- 2001
|
|
14
|
Hyperdynamic Importance Sampling
– Sminchisescu, Triggs
- 2002
|
|
14
|
Face-tracking and coding for video compression
– Vieux, Schwerdt, et al.
- 1999
|
|
13
|
Tracking multiple objects using the condensation algorithm
– Koller-Meier, Ade
- 2001
|
