Model-based object tracking in cluttered scenes with occlusions

We propose an efficient method for tracking 3D modelled objects in cluttered scenes. Rather than tracking objects in the image, our approach relies on the object recognition aspect of tracking. Candidate matches between image and model features define volumes in the space of transformations. The volumes of the pose space satisfying the maximum number of correspondences are those that best align the model with the image. Object motion defines a trajectory in the pose space. We give some results showing that the presented method allows tracking of objects even when they are totally occluded for a short while, without supposing any motion model and with a low computational cost (below 200 ms per frame on a basic workstation). Furthermore, this algorithm can also be used to initialize the tracking.

[1]  George C. Stockman,et al.  Matching Images to Models for Registration and Object Detection via Clustering , 1982, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  H. Baird.  Model-Based Image Matching Using Location , 1985 .

[3]  W. Eric L. Grimson,et al.  On the Sensitivity of the Hough Transform for Object Recognition , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Rachid Deriche,et al.  Tracking line segments , 1990, Image Vis. Comput..

[5]  D. W. Thompson,et al.  Three-dimensional model matching from an unconstrained viewpoint , 1987, Proceedings. 1987 IEEE International Conference on Robotics and Automation.

[6]  Rama Chellappa,et al.  A new approach to image feature detection with applications , 1996, Pattern Recognit..

[7]  Frans C. A. Groen,et al.  3D object recognition from 2D images using geometric hashing , 1992, Pattern Recognit. Lett..

[8]  Daniel DeMenthon.  De la vision artificielle à la réalité synthétique : système d'interaction avec un ordinateur utilisant l'analyse d'images vidéo. (From artificial vision to synthetic reality : system for interaction with a computer using video images analysis) , 1993 .

[9]  Todd A. Cass,et al.  Polynomial-Time Object Recognition in the Presence of Clutter, Occlusion, and Uncertainty , 1992, ECCV.

[10]  Daniel P. Huttenlocher,et al.  Measuring the Quality of Hypotheses in Model-Based Recognition , 1992, ECCV.

[11]  Thomas M. Breuel,et al.  Fast recognition using adaptive subdivisions of transformation space , 1992, Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[12]  Todd A. Cass.  Feature matching for object localization in the presence of uncertainty , 1990, [1990] Proceedings Third International Conference on Computer Vision.