Early vision and image analysis |
Visual processing starts from
measurements (feature extraction
and representation) on which image
operators are applied to extract shape
and motion primitives, texture
and color information. In this
phase the active vision paradigm
plays a relevant role in image
segmentation and grouping as
well in image and video
sequence analysis. |
Pattern recognition and image
inference |
Higher level of visual processing
regards statistical, structural
and syntactic pattern recognition.
Neural networks are considered
useful tools for learning and classification.
Genetic algorithms are becoming of
great interest in the search of global
solutions. Model acquisition, digital
geometry and shape reconstruction
techniques are important for 2D
and 3D object recognition. Spatial
reasoning needs new spatial data
structure and flexible inference
rules. |
Visual processing for
communication |
Multimedia databases, digital
and video libraries need efficient
image and video methods for
compression and coding. Image
databases and video processing
require suitable indexing and retrieval
solutions. Visual languages allows
the natural and optimized formulation of visual
computing. Both visual
communication and human-computer
interaction are based on advanced
visual interfaces. |
Applications |
Wide spectra of image analysis
applications are welcome (astronomy,
biology and biomedicine, cultural
heritage, OCR and document
analysis, mobile robots and visual
navigation, remote sensing and GIS,
surveillance, smart sensors and dedicated
architectures, visual inspection and quality
control,...). More emphasis will be
done to real cases and working
prototypes. |