A Generic Methodology for Partitioning Unorganised 3D Point Clouds for Robotic Vision

Authors:
Affiliations:
Venue:
CRV '04 Proceedings of the 1st Canadian Conference on Computer and Robot Vision
Year:
2004

Citing 6
Cited 4

Unsupervised Optimal Fuzzy Clustering

IEEE Transactions on Pattern Analysis and Machine Intelligence
Three-dimensional alpha shapes

ACM Transactions on Graphics (TOG)
Fast segmentation of range images into planar regions by scan line grouping

Machine Vision and Applications
An Experimental Comparison of Range Image Segmentation Algorithms

IEEE Transactions on Pattern Analysis and Machine Intelligence
MIR: An Approach to Robust Clustering-Application to Range Image Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Stochastic Algorithm for 3D Scene Segmentation and Reconstruction

ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part III

Morphological mesh filtering and α-objects

Pattern Recognition Letters
Affine-invariant contours recognition using an incremental hybrid learning approach

Pattern Recognition Letters
Incremental hybrid approach for unsupervised classification: applications to visual landmarks recognition

ICIAR'10 Proceedings of the 7th international conference on Image Analysis and Recognition - Volume Part I
Novelty detection and segmentation based on Gaussian mixture models: A case study in 3D robotic laser mapping

Robotics and Autonomous Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Range image segmentation has manyapplications in computer vision areas such as computergraphics and robotic vision. A generic methodologyfor 3D point set analysis in which planar structuresplay an important role is defined. It consistsmainly of a specific K-means algorithm whichis able to process different shapes in cluster. At thesame time, within geometric and topologic considerations,a set of application-driven heuristics is designed.This helps to find out the right number ofstructures in point sets in order to give a good visualizationand representation of a large scale environmentwithout a priori models. Our aim is topropose a simple and generic frame for 3D scene understanding.Tests were realised on different types ofenvironment data: natural and man-made. This researchproject has been realized with EADS (FrenchAir Space Society).