Abstract. In this paper, we introduce background cut, a high quality and realtime foreground layer extraction algorithm. From a single video sequence with a moving foreground objec...
We show how a special decomposition of a set of two or three general projection matrices, called canonic enables us to build geometric descriptions for a system of cameras which a...
We present a background model that differentiates between background motion and foreground objects. Unlike most models that represent the variability of pixel intensity at a partic...
Morphological operators designed for grey-scale functions process every points of the space identically whatever their luminance. In many situations however, it is interesting to ...
A component framework for defining content-based network processing is presented. It is unique among other existing approaches because it enables contents to naturally define th...