Spatiotemporal segmentation and tracking of objects for visualization of videoconference image sequences

In this paper, a procedure is described for the segmentation, content-based coding, and visualization of videoconference image sequences. First, image sequence analysis is used to estimate the shape and motion parameters of t he person facing the camera. A spatiotemporal filter, taking into account t he intensity differences between consequent frames, is applied, in order to separate the moving person from the static background. The foreground is s egmented in a number of regions in order to identify the face. For this pur pose, we propose the novel procedure of K-Means with connectivity constrain t algorithm as a general segmentation algorithm combining several types of information including intensity, motion and compactness. In this algorithm, the use of spatiotemporal regions is introduced since a number of frames a re analyzed simultaneously and as a result, the same region is present in c onsequent frames. Based on this information, a 3-D ellipsoid is adapted to the person's face using an efficient and robust algorithm. The rigid 3-D mo tion is estimated next using a least median of squares approach. Finally, a Virtual Reality Modeling Language (VRML) file is created containing all th e above information; this file may he viewed by using any VRML 2.0 complian t browser.