This paper deals with the problem of analyzing and visualizing large size volume data sets. To this aim, we define a three-dimensional multiresolution model based on unstructured tetrahedral meshes, and on a vertex- insertion/removal simplification strategy, that we call a Vertex-based Multi-Tessellation (MT). We propose a new compact data structure for a vertex-based MT, and we analyze it with respect to both its space requirements and its efficiency in supporting primitives needed for implementing selective refinement. We show that this data structure provides very good compression ratios not only with respect to a data structure which encodes a general-purpose three-dimensional Multi-Tessellation, but also with respect to encoding the original mesh at full resolution. We also compare it with a data structure for an edge-based Multi-Tessellation both in terms of storage and on the performances in a selective refinement query.
