The analysis of 3D scenes consisting of nonrigid moving objects from 2D image sequences is discussed. A parametric description of dynamic objects is extracted and the time variant scene parameters are estimated throughout the sequence by employing an analysis by synthesis approach. From the parametric scene description images are synthesized and compared with the original input images of the camera. Frame differences between both images are evaluated to estimate a scene parameter update. The analysis system is applied to video phone scenes as a data compression algorithm where the scene parameters are transmitted and the output sequence is synthesized at the receiver. Index Terms -- 3D scene analysis, dynamic scene analysis, image sequence analysis, 3D motion estimation, 3D object tracking, 3D shape reconstruction, 3D shape adaptation, analysis by synthesis, object--oriented video phone coding. I. INTRODUCTION Modelling of 3D scenes from 2D image sequences has been a research topic for...

An algorithm for estimating reliable and accurate depth maps from stereoscopic image pairs is presented, which is based on block--matching techniques for disparity estimation. By taking neighboring disparity values into account, reliability and accuracy of the estimated disparity values are increased and the corona effect at disparity discontinuities is avoided. An interpolation of disparity values within segmented regions of homogeneous disparity enables the computation of dense depth maps by means of triangulation. 1 Introduction Depth estimation is used in applications like 3D--modelling of natural objects [1] [2], 3D--remote handling and quality control [3]. Depth information is obtained by a triangulation of corresponding image points with known stereoscopic camera parameters. Therefore, the coordinate difference between corresponding image points, called disparity, has to be estimated. Applying common block--matching techniques for disparity estimation, the correspondence of i...

We introduce a compact coding of image information which explicitely separates geometric information (orientation) and structural information (phase and color). We investigate the importance of these factors for stereo matching on a large data set. From these investigation we can conclude that it is their combination that gives the best results. Concrete

In this paper we address the statistics of second order relations of feature vectors derived from image sequences. We compute the individual vector components corresponding to the visual modalities orientation, contrast transition, optic flow, and color by conventional lowlevel early vision algorithms. As a main result, we observe that collinear (or parallel) line pairs are with very great likelihood also associated with other identical features, for example sharing the same flow pattern, or color or even sharing multiple feature combinations. It is known

A knowledge based approach for the automatic generation of 3D models from stereo images is presented. Two different application fields are shown: The modelling of houses and the modelling of landscapes from aerial images. The models are used for computer graphic applications. A human observer is in possession of a priori knowledge about the scene. To meet his expectations and thus to improve realism the presented approach employs explicit knowledge. The knowledge base is exploited to derive constraints for surface reconstruction and to complete partially occluded models. 1 Introduction For visualization of synthetic scenes 3D models are required from which new simulated views can be computed. Applications such as flight and driving simulators, architecture and landscape visualization and film industry have a high demand for realistic models. Quantity, precision and the kind of models ask for methods that automate the model generation of real objects and are applicable to outdoor scene...

Visualization of spatial data in a virtual reality environment is catching the imagination of researchers across the disciplines of spatial information sciences and computer vision alike. Different approaches have been investigated into for visualization of three-dimensional objects in a virtual reality environment. An important aspect of creation of virtual world is the increasing interest shown by the research community in building virtual world database with contents of real world object. Photogrammetric techniques have become an important link in the process of creation of virtual worlds.

In this paper we address the problem of predetermined structural knowledge human and artificial visual systems have to apply to be able to start a bootstrapping process that enables them to interact with the world efficiently. We focus on two different kind of regularities, namely deterministic and statistical, which visual system need to apply to overcome the ambiguity of visual low level processes.

This paper discusses the concept of maturity in the use of GIS and then formulates a computational method for measuring an organisations maturity level from the construction of a surrogate indicator. Generation of this model is made under the proposition that maturity is linked to the level that GIS has been integrated and utilised on an organisation wide basis in day to day activities. The research focuses on New Zealand local government and incorporates parallel studies of conventional information technology (IT) with recently collected data to provide support for the concepts and techniques used. It is postulated that due to similarities of function found in other local authorities, that the model has the potential, with further research for wide application.

This paper describes a texture based shape connection method for reconstructing parts of an object shape into its entire 3-D shape. This method employs a stereo camera to obtain the parts of an object. First, this method finds overlaps between the parts, obtained from a number of different viewpoints, using texture. Then, the method uses the results for estimating the relative position-and-orientation between the parts. Finally, it reconstructs the entire shape by connecting the parts. The experimental results show that this method yields accurate reconstruction results, even when an object lacks shape features.