DETECTING CONTOURS IN IMAGE SEQUENCES

This paper describes a new algorithm for finding the contours of a mov ing object in an image sequence. A distinctive feature of this algorit hm is its complete bottom-up strategy from image data to a consistent contour description. In our algorithm, an input image sequence is imme diately converted to a complete set of quasi logical spatio-temporal m easures on each pixel, which provide constraints on varying brightness . Then, candidate regions in which to localize the contour are bounded based on consistent grouping among neighboring measures. This reduces drastically the ambiguity of contour location. Finally, some mid-leve l constraints on spatial and temporal smoothness of moving boundaries are invoked, and they are combined with these low-level measures in th e candidate regions. This is performed efficiently by the regularizati on over the restricted trajectory of the moving boundary in the candid ate regions. Since any quantity is dimensionless, the results are not affected by varying conditions of camera and objects. We examine the e fficiency of this algorithm through several experiments on real NTSC m otion pictures with dynamic background and natural textures.