Multisensor Fusion for Computer Vision

Fusion or integration of information from multiple sensors allows a computer to achieve the optimum interpretation of a visual scene. This NATO workshop volume presents current research in the field by the leading research groups in Europe and North America.

Information Integration and Model Selection in Computer Vision.- Principles and Techniques for Sensor Data Fusion.- The Issues, Analysis, and Interpretation of Multisensor Images.- Physically-Based Fusion of Visual Data over Space, Time, and Scale.- What Can Be Fused?.- Kalman Filter-Based Algorithms for Estimating Depth from Image Sequences.- Robust Linear Rules for Nonlinear Systems.- Geometric Sensor Fusion in Robotics (Abstract).- Cooperation between 3D Motion Estimation and Token Trackers (Abstract).- Three-Dimensional Fusion from a Monocular Sequence of Images.- Fusion of Range and Intensity Image Data for Recognition of 3D Object Surfaces.- Integrating Driving Model and Depth for Identification of Partially Occluded 3D Models.- Fusion of Color and Geometric Information.- Evidence Fusion Using Constraint Satisfaction Networks.- Multisensor Information Integration for Object Identification.- Distributing Inferential Activity for Synchronic and Diachronic Data Fusion.- Real-Time Perception Architectures: The SKIDS Project.- Algorithms on a SIMD Processor Array.- Shape and Curvature Data Fusion by Conductivity Analysis (Abstract).- A Knowledge-Based Sensor Fusion Editor.- Multisensor Change Detection for Surveillance Applications.- Multisensor Techniques for Space Robotics.- Coordinated Use of Multiple Sensors in a Robotic Workcell.- Neural Network Based Inspection of Machined Surfaces Using Multiple Sensors.- Adaptive Visual/Auditory Fusion in the Target Localization System of the Barn Owl.- Index of Key Terms.- Workshop Speakers.