MEPP2 Project
Compression Valence Filter

Overview

The Compression Valence filter implements a progressive compression for 3D meshes.
The progressive compression method simplifies iteratively the input mesh to generate differents levels of details (LoDs).
These LoDs can then be transmitted progressively in a coarse-to-fine way at the decompression.
In particular, our method adapts the quantization precision (both geometry and color) to each LoD in order to optimize the rate-distortion (R-D) trade-off.

Our joint progressive compression and reversible watermarking method (not implemented here) offers a possibility to embed an watermark information in order to protect the ownership of the input mesh.

See the Decompression Valence Filter for the decompression step.

Parameters

Parameter Description
Compressed file name Name for the compressed file. The file extension should be ".p3d".
With adaptative quantization Enable or disable the use of the adaptation of quantization precision.
Quantization bits Number of bits for the geometry quantization.
Max vertices Maximum number of vertices of the final (the coarsest) mesh.

Supported data structures

The supported data structures are: CGAL Polyhedron3, CGAL Surface_mesh, CGAL LCC, OpenMesh. AIF is not yet supported.

Restrictions

Applies to manifold triangular meshes only.

References

Progressive compression of colored meshes

H. LEE, G. LAVOUE, F. DUPONT,
Rate-distortion optimization for progressive compression of 3D mesh with lor attributes,
The Visual Computer, 2011.

Joint watermarking and progressive compression of 3D meshes

H. LEE, C. DIKICI, G. LAVOUE, F. DUPONT,
Joint reversible watermarking and progressive compression of 3D meshes,
The Visual Computer 27(6-8):781-792, 2011.