![](bg8.png)
An interesting remark can be made based on the re-
sults of the evaluation. Two of the evaluators sys-
tematically gave a lower grade to the data-driven al-
gorithm, for every model at every evaluated bitrate.
Even in situations where all of the other evaluators
gave a significant advantage to the data-driven al-
gorithm, these 2 evaluators graded it lower. We
could therefore think that some individuals are or be-
come more sensitive to geometry distortion than oth-
ers. These two subjective experiments have shown
that driving the compression with attributes like color
can improve the rate-quality compromise, whereas
producing a slighty higher geometric distorsion. Of
course this improvement depends on the geometric
structure of the object and on the importance of the
attributes in its visual appearance: we have noticed
that it is more interesting to use our compression tech-
nique on objects where the important data is color,
instead of meshes where the geometry-color tradeoff
can dramatically deteriorate the overall visual quality.
5 CONCLUSIONS
We have presented a framework for the data-driven
progressive compression of meshes. It allows the use
of any associated property, and the behavior of the
compression process can be customized through a sin-
gle distance function per property. We showed an ex-
ample of color data-driving and compression of the
color property. At equal bitrates, it improves the vi-
sual quality of the decompressed meshes compared to
the non data-driven version. This algorithm is well
suited for the compression of meshes where geome-
try and topology play a secondary role compared to
associated properties, such as with scientific visual-
ization models. This work sets the bases for further
research around the concept of data-driven compres-
sion. The performance of the algorithm relies in part
on efficient distance functions. Other properties will
be considered for testing, such as normals. More
complex and efficient distance functions could be de-
fined for intrinsic properties such as curvature tensors
and connectivity data, in order to improve the geo-
metric shape of the intermediate meshes. Moreover,
a BSP-tree could be used to allow more flexible cuts.
However, special attention would have to be given to
plane coding and overall processing time.
ACKNOWLEDGEMENTS
This work has been supported by French National Re-
search Agency (ANR) through SCOS project (ANR-
06-TLOG-029) and COSINUS program (project
COLLAVIZ ANR-08-COSI-003). The authors would
like to thank R. Marc and C. Mouton from EDF for
their help regarding the perceptive evaluation.
REFERENCES
Alliez, P. and Desbrun, M. (2001). Valence-Driven connec-
tivity encoding for 3D meshes. Computer Graphics
Forum, 20:480–489.
Bordignon, A., Lewiner, T., Lopes, H., Tavares, G., and
Castro, R. (2006). Point set compression through BSP
quantization. In Proceedings of the Brazilian Sympo-
sium on Computer Graphics and Image Processing,
pages 229–238.
Cignoni, P., Rocchini, C., and Scopigno, R. (1998). Metro:
measuring error on simplified surfaces. Computer
Graphics Forum, 17:167–174.
Gandoin, P. and Devillers, O. (2002). Progressive loss-
less compression of arbitrary simplicial complexes.
In SIGGRAPH, pages 372–379, San Antonio, Texas.
ACM.
Hoppe, H. (1996). Progressive meshes. In SIGGRAPH,
pages 99–108. ACM.
Huang, Y., Peng, J., Kuo, C. C. J., and Gopi, M. (2008).
A generic scheme for progressive point cloud cod-
ing. IEEE Transactions on Visualization and Com-
puter Graphics, 14(2):440–453.
Karni, Z. and Gotsman, C. (2000). Spectral compression of
mesh geometry. In SIGGRAPH, pages 279–286. ACM
Press/Addison-Wesley Publishing Co.
Khodakovsky, A., Schr
¨
oder, P., and Sweldens, W. (2000).
Progressive geometry compression. In SIGGRAPH,
pages 271–278. ACM Press/Addison-Wesley Publish-
ing Co.
Peng, J., Eckstein, I., and Kuo, C. J. (2006). A novel
and efficient progressive lossless mesh coder. In SIG-
GRAPH Sketches, page 180, Boston, Massachusetts.
ACM.
Peng, J., Kim, C., and Kuo, C. J. (2005). Technolo-
gies for 3D mesh compression: A survey. Journal
of Visual Communication and Image Representation,
16(6):688–733.
Peng, J. and Kuo, C. J. (2005). Geometry-guided progres-
sive lossless 3D mesh coding with octree (OT) decom-
position. ACM Transactions on Graphics, 24(3):609–
616.
Shilane, P., Min, P., Kazhdan, M., and Funkhouser, T.
(2004). The princeton shape benchmark. In Proceed-
ings of the Shape Modeling International, pages 167–
178. IEEE Computer Society.
Taubin, G. and Rossignac, J. (1998). Geometric compres-
sion through topological surgery. ACM Transactions
on Graphics, 17(2):84–115.
Waschbsch, M., Gross, M., Eberhard, F., Lamboray, E., and
Wrmlin, S. (2004). Progressive compression of Point-
Sampled models. Proceedings of the Eurographics
Symposium on Point Based Graphics, pages 95–102.
GRAPP 2010 - International Conference on Computer Graphics Theory and Applications
12