Issues in acquiring, processing and visualizing large and detailed 3D models

Modelling from reality using active optical geometric sensing has been a very active research area in computer graphics and vision for the last twenty years. While most elements of the modelling pipeline have reached maturity and have been adopted in several application sectors, several issues remain, particularly in the modelling of large structures and environments, as well as in the management of large, complex and detailed 3D models. This paper describes some of these issues, and outlines some of the solutions that we have proposed. These methods and approaches, as well as their current limitations, are described using different example applications: a monument (the Erechtheion), a painting (Mona Lisa), and a terrain model.