In the above areas, Gian Luca Foresti conducts and supervises research
work within the frameworks of some National and International projects.
Intruder detection
Human behaviour analysis and understanding
Event detection Real time monitoring of outdoor enviroments (railway stations, airports, motorway, etc...)
Face detection and recognition
Mobile robot navigation for indoor surveillance
Spatial and temporal analysis of video sequences
Motion detection and estimation with fixed and mobile camera
Robust object tracking
Shadow detection and elimination
Distributed networks of Extended Kalman Filters.
Multisensor image processing Top
Multisensor (range, infrared, optical) image analysis
Low-level image processing: image acquisition, pre-processing, enhancement
and filtering
Change-detection and attention-focusing methods (blob extraction)
Linear and non-linear feature extraction and grouping (2D and 3D geometric
primitives)
Perceptual grouping and image segmentation (edges and regions)
Object detection and tracking with a mobile camera (PTZ or placed on autonomous
robot)
Robust feature tracking
Outlier feature detection and rejection
Ego-motion estimation and compensation
Automatic regulation of camera parameters (focus, iris, zoom)
3D scene understanding with variable illumination conditions
Integration of multisensor data (optical, infrared, radar, etc...)
Distributed sensory networks (DSNs)
Bayesian Networks
Multilevel Markov Random Fields.
Real-time recognition of 2D and 3D objects in real environments
Neural Networks (multilayer perceptron, neural trees, self organizing maps)
Decision trees and associative memories
Template matching.