Statistical Analysis of the Global Geodesic Function for 3D Object Classification

Djamila Aouada, Shuo Feng, Hamid Krim

Abstract

Thispaper presents anovel classification strategy for3Dobjects. Ourtechnique isbased onusing aGlobal Geodesic Function tointrinsically describe thesurface ofanobject. The choice oftheGlobal Geodesic Function ensures theinvarianceoftheclassification procedure toscaling andall isometrictransformations. UsingtheJensen-Shannon Divergence, feature parameters areextracted fromtheprobability distribution functions oftheGlobal Geodesic Function foreach oneoftheclasses. These parameters areusedinthedecision ofaclass membership ofanobject. Thisapproach demonstrates lowcomputational cost, efficiency, androbustness to resolution overmanydifferent data sets.
BibTeX
@inproceedings{aouada2007statistical,
  author = {Djamila Aouada and Shuo Feng and Hamid Krim},
  title = {{Statistical Analysis of the Global Geodesic Function for 3D Object Classification}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2007},
  doi = {10.1109/icassp.2007.365990},
}

Loading related content...

Figure