A selective vision system sequentially collects evidence to support a specified hypothesis about a scene, as long as the additional evidence is worth the effort of obtaining it. Efficiency comes from processing the scene only where necessary, to the level of detail necessary, and with only the necessary operators. Knowledge representation and sequential decision-making are central issues for selective vision, which takes advantage of prior knowledge of a domain's abstract and geometrical structure and models for the expected performance and cost of visual operators. The TEA-1 selective vision system uses Bayes nets for representation and benefitcost analysis for control of visual and non-visual actions. It is the high-level control for an active vision system, enabling purposive behavior, the use of qualitative vision modules and a pointable multiresolution sensor. TEA-1 demonstrates that Bayes nets and decision theoretic techniques provide a general, re-usable framework for constructi...

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...

A stereoscopic scene analysis system for 3–D modeling of objects from stereoscopic image sequences is described. A dense map of 3–D surface points is obtained by image correspondence, object segmentation, interpolation, and triangulation. Emphasis is put on the accurate measurement of image correspondences from grey level images. The surface geometry of each scene object is approximated by a triangular wire–frame which stores the surface texture in texture maps. Sequence processing serves to track camera motion and to fuse surfaces from different view points into a consistent 3–D surface model. From the textured 3–D models, highly realistic image sequences from arbitrary view points can be synthesized using computer graphics techniques. 1

Two paradigms for visual analysis are top-down, starting from high-level models or information about the image, and bottom-up, where little is assumed about the image or objects in it. We explore a local, bottom-up approach to image analysis. We develop operators to identify and classify image junctions, whichcontain important visual cues for identifying occlusion, transparency, and surface bends. Like the human visual system, we begin with the application of linear filters which are oriented in all possible directions. Wedevelop an efficientway to create an oriented filter of arbitrary orientation by describing it as a linear combination of basis filters. This approach to oriented filtering, which we call steerable filters, offers advantages for analysis as well as computation. We design a variety of steerable filters, including steerable quadrature pairs, which measure local energy. We show applications of these filters in orientation and texture analysis, and image representation and enhanc...

Introduction The rapid progress in the development of powerful computer graphics hardware and software enables users in a wide range of applications to gain a better insight into processes by visual simulation. Suppliers of flight and driving simulators as well as landscape and city planners are interested to simulate photo--realistic views of the environment. Architects and city planners for example construct new buildings with CAD systems and are interested to visualize their impact onto the existing environment beforehand. Complete realism, however, is possible only if the buildings to be constructed are placed inside a 3--D reconstruction of of the real environment. It is therefore necessary to reconstruct the existing environment as a 3--D model of the real scene with as little effort as possible [1]. One possible approach is to obtain a complete 3--D scene description by evaluating images of the scene. Modeling of 3--D scenes from 2D image sequences has been a research t

OF THE DISSERTATION Composition of Image Analysis Processes through Object-Centered Hierarchical Planning by Leiguang Gong, Ph.D. Dissertation Director: Professor Casimir A. Kulikowski The present thesis describes a new approach to biomedical image interpretation --- the knowledge-based composition of image analysis processes through object-centered hierarchical planning. Its computer implementation in the VISIPLAN system has been tested on different image analysis tasks using Magnetic Resonance (MR) scans. The approach has been shown to be flexible and effective in handling a reasonably broad class of image analysis problems. The new approach defines the problem of composing image analysis processes primarily as that of detecting a sequence of reference objects to establish the context for segmentation and recognition of a goal object at the highest level of problem solving. At a lower level it defines the composition of specific analysis processes, which are scheduled for each of the...

( * at the time of this paper working at Department of Electrical and Electronic Engineering,

A vision–based 3–D scene analysis system is described that is capable to model complex real–world scenes like buildings automatically from stereoscopic image pairs. Input to the system is a sequence of stereoscopic images taken with two standard CCD Cameras and TV lenses. The relative orientation of both cameras to each other is known by calibration. The camera pair is then moved throughout the scene and a long sequence of closely spaced views is recorded. Each of the stereoscopic image pairs is rectified and a dense map of 3–D surface points is obtained by area correlation, object segmentation, interpolation, and triangulation. 3–D camera motion relative to the scene coordinate system is tracked directly from the image sequence which allows to fuse 3–D surface measurements from different viewpoints into a consistent 3–D model scene. The surface geometry of each scene object is approximated by a triangular surface mesh which stores the surface texture in a texture map. From the textured 3–D models, realistic looking image sequences from arbitrary view points can be synthesized using computer graphics.

In this paper we describe a real time object tracking system consisting of three modules (motion detection, object tracking, robot control), each working with a moderate accuracy, implemented in parallel on a workstation cluster, and therefore operating fast without any specialized hardware. The robustness and quality of the system is achieved by a combination of these vision modules with an additional attention module which recognizes errors during the tracking. For object tracking in image sequences we apply the method of active contour models (snakes) which can be used for contour description and extraction as well. We show how the snake is initialized automatically by the motion detection module, explain the tracking module, and demonstrate the detection of errors during the tracking by the attention module. Experiments show that this approach allows a robust real--time object tracking over long image sequences. Using a formal error measurement presented in this paper it will be sh...

Checkers, a complex real-time application, demonstrates the advantages of decomposing animate vision systems by function and independently selecting an appropriate ' parallel-programming model for each function. v ision can be viewed as a passive, observational activity, or as one intimate ly related to action (for example, manipulation, navigation). In passive