With the great advances in the fields of telecommunication, graphic hardware and geometry processing, the vizualisation and manipulation of 3D large data sets are now commonplace in the industry, and particularly for scientific simulation.
One of the critical issue is to be able to share the visualization and the manipulation of this 3D content between several users (i.e. workstations). This kind of collaborative visualization services have to handle variable bandwidth networks and visualization clients with different hardware resources (thin or rich clients).
In this context, there is an increasing demand for:
reducing the size of this 3D content in order to permit a low bandwith transmission.
reducing the complexity of this 3D content, in order to adapt to thin clients like mobile terminals.
providing progressive transmission which is necessary to enable real time interaction, analysis and manipulation of the 3D data.

The reducing of the size and the complexity of these meshes consists in suppressing the intrinsic data rendundancy but also reducing the number of vertices / faces within a multi-resolution framework.
Another important aspect in the sharing and transmission of these large data sets concerns the protection and the integrity of the data. Digital watermarking has been considered as a potential efficient solution for copyright protection of various multimedia contents.

Even if many 3D mesh compression methods have been proposed in the literature very few of them are able to handle these attached attributes, whereas they can represent a huge amount of data, compared to the geometry. Hence one of the objectives of this PhD thesis is to provide new progressive methods, allowing compression of both the geometry and the attributes, in a multi-resolution framework: a low resolution coarse model is first transmitted followed by iterative refinements.
The protection of 3D models, through watermarking has also been extensively studied in the scientific community. However not any existing method is able to survive to progressive compression. Hence our objective is to insert the protection step (i.e. the watermarking) within the compression process.