Résumé : Visual sensors provide exclusively uncertain and partial knowledge of a scene. In this report, we present a suitable scene knowledge representation that makes integration and fusion of new, uncertain and partial sensor measures possible. It is based on a mixture of stochastic and set membership models. We consider that, for a large class of applications, an approximated representation is sufficient to build a preliminary map of the scene. Our approximation mainly results in ellipsoidal calculus by means of a normal assumption for stochastic laws and ellipsoidal over or inner bounding for uniform laws. With these approximations, we coarsely model objects by their including ellipsoid. Then we build an efficient estimation process integrating visual data online in order to refine the location and approximated shape of the objects. Based on this estimation scheme, we perform online and optimal exploratory motions for the camera.