Logo image
Back
RoPS: A local feature descriptor for 3D rigid objects based on rotational projection statistics
Conference paper

RoPS: A local feature descriptor for 3D rigid objects based on rotational projection statistics

Y. Guo, F.A. Sohel, M. Bennamoun, J. Wan and M. Lu
2013 1st International Conference on Communications, Signal Processing, and their Applications (ICCSPA), pp.1-6
1st International Communications Conference on Signal Processing, and their Applications (ICCSPA) 2013 (Sharjah, 12/02/2013–14/02/2013)
2013
DOI: https://doi.org/10.1109/ICCSPA.2013.6487310
url
Link to Published Version *Subscription may be requiredView

Abstract

The proper choice of local surface feature descriptors is a key step for an accurate and robust surface matching between different range images. This paper presents a novel 3D feature descriptor for free form objects based on rotational projection statistics. A rotation invariant local reference frame for each feature point is defined by performing an eigenvalue decomposition on the covariance matrix formed by all points lying on the local surface. The feature descriptor is then constructed by rotationally projecting the neighboring 3D points onto 2D planes and by calculating low order moments and the entropy of the 2D distribution matrix on these planes. Experiments were performed on a dataset comprised of 45 scenes, and the results show that the proposed method is robust to noise and variations in mesh resolution.

Details

Metrics

33 Record Views
Logo image