Statistical mosaics for tracking

Abstract A method of robust feature-detection is proposed for visual tracking with a pan-tilt head. Even with good foreground models, the tracking process is liable to be disrupted by strong features in the background. Previous researchers have shown that the disruption can be somewhat suppressed by the use of image-subtraction. Building on this idea, a more powerful statistical model of background intensity is proposed in which a Gaussian mixture distribution is fitted to each of the pixels on a ‘virtual’ image plane. A fitting algorithm of the ‘Expectation-Maximisation’ type proves to be particularly effective here. Practical tests with contour tracking show marked improvement over image subtraction methods. Since the burden of computation is off-line, the online tracking process can run in real-time, at video field-rate.

[1]  Laurent D. Cohen,et al.  A finite element method applied to new active contour models and 3D reconstruction from cross sections , 1990, [1990] Proceedings Third International Conference on Computer Vision.

[2]  Jitendra Malik,et al.  Robust Multiple Car Tracking with Occlusion Reasoning , 1994, ECCV.

[3]  G. D. Sullivan,et al.  Natural and artificial low-level seeing systems - Visual interpretation of known objects in constrained scenes , 1992 .

[4]  Richard Szeliski,et al.  Tracking with Kalman snakes , 1993 .

[5]  Anup Basu,et al.  Motion Tracking with an Active Camera , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Nicholas Ayache,et al.  Medical image tracking , 1993 .

[7]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[8]  Andrew Blake,et al.  Real-time Visual Tracking for Surveillance and Path Planning , 1992, ECCV.

[9]  Ulf Grenander,et al.  Hands: A Pattern Theoretic Study of Biological Shapes , 1990 .

[10]  Jan-Olof Eklundh,et al.  Computer Vision — ECCV '94 , 1994, Lecture Notes in Computer Science.

[11]  Timothy F. Cootes,et al.  Building and using flexible models using grey-level information , 1993, ICCV 1993.

[12]  G. Sandini,et al.  Computer Vision — ECCV'92 , 1992, Lecture Notes in Computer Science.

[13]  Ian Craw,et al.  Finding Image Features Using Deformable Templates and Detailed Prior Statistical Knowledge , 1991, BMVC.

[14]  D B Arnold,et al.  A neural network model of the vestibulo-ocular reflex using a local synaptic learning rule. , 1992, Philosophical transactions of the Royal Society of London. Series B, Biological sciences.

[15]  Andrew Blake,et al.  The dynamic analysis of apparent contours , 1990, [1990] Proceedings Third International Conference on Computer Vision.

[16]  Arthur Gelb,et al.  Applied Optimal Estimation , 1974 .

[17]  David C. Hogg,et al.  Learning Flexible Models from Image Sequences , 1994, ECCV.

[18]  Dana H. Ballard,et al.  Computer Vision , 1982 .

[19]  Christopher M. Brown,et al.  Real-time smooth pursuit tracking , 1993 .

[20]  Martin A. Fischler,et al.  The Representation and Matching of Pictorial Structures , 1973, IEEE Transactions on Computers.

[21]  Michael Brady,et al.  Design of stereo heads , 1993 .

[22]  Alan L. Yuille,et al.  Deformable templates , 1993 .

[23]  Richard Szeliski,et al.  Image mosaicing for tele-reality applications , 1994, Proceedings of 1994 IEEE Workshop on Applications of Computer Vision.

[24]  Astrom.  Computer Controlled Systems , 1990 .

[25]  Chris Harris,et al.  Tracking with rigid models , 1993 .

[26]  David W. Murray,et al.  Active Camera Calibration for a Head-Eye Platform Using the Variable State-Dimension Filter , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[27]  James J. Clark,et al.  Modal Control Of An Attentive Vision System , 1988, [1988 Proceedings] Second International Conference on Computer Vision.