Keypoints labeling for backgroung substraction in tracking applications

This paper studies the problem of background/object differentiation in a keypoint-based tracking application where the object is delimited with a bounding box. We present a keypoint labeling algorithm based on four features: the label of the matched keypoint, color, motion, and position. We discuss methods to best exploit these features, then we detail our labeling algorithm and validate it with some experiments on several tracking video sequences.

[1]  Touradj Ebrahimi,et al.  Tracking video objects in cluttered background , 2005, IEEE Transactions on Circuits and Systems for Video Technology.

[2]  Cordelia Schmid,et al.  Indexation à l'aide de points d'intérêt invariants à l'échelle , 2001 .

[3]  Bernard Mérialdo,et al.  Generic object tracking for fast video annotation , 2007, VISAPP.

[4]  Charles V. Stewart,et al.  Robust Parameter Estimation in Computer Vision , 1999, SIAM Rev..

[5]  Axel Techmer,et al.  Contour-based motion estimation and object tracking for real-time applications , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[6]  Christopher G. Harris,et al.  A Combined Corner and Edge Detector , 1988, Alvey Vision Conference.

[7]  Rachid Deriche,et al.  Differential invariants for color images , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[8]  Rémy Prost,et al.  Mesh-based video objects tracking combining motion and luminance discontinuities criteria , 2004, Signal Process..

[9]  Y. Bar-Shalom.  Tracking and data association , 1988 .

[10]  Zhengyou Zhang,et al.  Parameter estimation techniques: a tutorial with application to conic fitting , 1997, Image Vis. Comput..

[11]  T. Remi,et al.  Probabilistic Matching Algorithm for Keypoint Based Object Tracking Using a Delaunay Triangulation , 2007, Eighth International Workshop on Image Analysis for Multimedia Interactive Services (WIAMIS '07).