This paper presents a new method for structuring and tracking of objects in video sequences. Our approach is based on image moments and the bsp-tree data structure. We use invariant properties of these moments to construct a bsp-tree and determine an ellipsis that approximates the object's shape. Then, we employ this information to track objects frame by frame through the image sequence. The method works well for segmented images with a single object and we assume that the motion will not change abruptly.

No abstract

We present a two-step camera calibration process based on linear least squares formulations. We show how to apply this process to build an integrated modeling environment, in which a scene is modeled by having the user specify, on a given image, a set of reference points (used for camera calibration) and object points (used for positioning the objects in the scene). Once this has been done, synthetic information can be superimposed onto the image and arbitrary views can be produced. Using this methodology, we developed Virtual Referee, an application to analyze soccer plays from still images, generating an OpenGL camera over a 3D field model.

This article describes a method for object and terrain visualization by means of the combination of two algorithms, one for terrain data and one for objects. Our purpose is to generate, efficiently and rapidly, aerial images of terrain with obje1;ts such as houses, vehicles, and transmission lines, thus allowing a simulated flight. For the objects, described by lines and polygons, the Z-Buffer algorithm is used; for the terrain, described by height maps, an optimized Ray-Casting Algorithm, called Floating Horizon Algorithm, is used.