Improving 3D keypoint detection from noisy data using growing neural gas

Authors:
José Garcia-Rodriguez;Miguel Cazorla;Sergio Orts-Escolano;Vicente Morell
Affiliations:
Department of Computing Technology, University of Alicante, Spain;Instituto de Investigación en Informática, University of Alicante, Spain;Department of Computing Technology, University of Alicante, Spain;Instituto de Investigación en Informática, University of Alicante, Spain
Venue:
IWANN'13 Proceedings of the 12th international conference on Artificial Neural Networks: advences in computational intelligence - Volume Part II
Year:
2013

Citing 6
Cited 0

Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography

Communications of the ACM
An Affine Invariant Interest Point Detector

ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part I
Indoor-Outdoor Image Classification

CAIVD '98 Proceedings of the 1998 International Workshop on Content-Based Access of Image and Video Databases (CAIVD '98)
Distinctive Image Features from Scale-Invariant Keypoints

International Journal of Computer Vision
Three-dimensional surface reconstruction using meshing growing neural gas (MGNG)

The Visual Computer: International Journal of Computer Graphics
Fast point feature histograms (FPFH) for 3D registration

ICRA'09 Proceedings of the 2009 IEEE international conference on Robotics and Automation

Quantified Score

Hi-index	0.00

Visualization

Abstract

3D sensors provides valuable information for mobile robotic tasks like scene classification or object recognition, but these sensors often produce noisy data that makes impossible applying classical keypoint detection and feature extraction techniques. Therefore, noise removal and downsampling have become essential steps in 3D data processing. In this work, we propose the use of a 3D filtering and down-sampling technique based on a Growing Neural Gas (GNG) network. GNG method is able to deal with outliers presents in the input data. These features allows to represent 3D spaces, obtaining an induced Delaunay Triangulation of the input space. Experiments show how the state-of-the-art keypoint detectors improve their performance using GNG output representation as input data. Descriptors extracted on improved keypoints perform better matching in robotics applications as 3D scene registration.