Real-Time scale invariant 3d range point cloud registration

Authors:
Anuj Sehgal;Daniel Cernea;Milena Makaveeva
Affiliations:
Indian Underwater Robotics Society, Noida, India;Department of Computer Science, University of Kaiserslautern, Kaiserslautern, Germany;Computer Science, Jacobs University Bremen, Bremen, Germany
Venue:
ICIAR'10 Proceedings of the 7th international conference on Image Analysis and Recognition - Volume Part I
Year:
2010

Citing 8
Cited 0

Iterative point matching for registration of free-form curves and surfaces

International Journal of Computer Vision
Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography

Communications of the ACM
Object Recognition from Local Scale-Invariant Features

ICCV '99 Proceedings of the International Conference on Computer Vision-Volume 2 - Volume 2
Distinctive Image Features from Scale-Invariant Keypoints

International Journal of Computer Vision
GNU Scientific Library Reference Manual - 2nd Edition

GNU Scientific Library Reference Manual - 2nd Edition
Image-based registration of 3D-range data using feature surface elements

VAST'04 Proceedings of the 5th International conference on Virtual Reality, Archaeology and Intelligent Cultural Heritage
VCLab''s tools for 3D range data processing

VAST'03 Proceedings of the 4th International conference on Virtual Reality, Archaeology and Intelligent Cultural Heritage
Robust filtering of noisy scattered point data

SPBG'05 Proceedings of the Second Eurographics / IEEE VGTC conference on Point-Based Graphics

Quantified Score

Hi-index	0.00

Visualization

Abstract

Stereo cameras, laser rangers and other time-of-flight ranging devices are utilized with increasing frequency as they can provide information in the 3D plane. The ability to perform real-time registration of the 3D point clouds obtained from these sensors is important in many applications. However, the tasks of locating accurate and dependable correspondences between point clouds and registration can be quite slow. Furthermore, any algorithm must be robust against artifacts in 3D range data as sensor motion, reflection and refraction are commonplace. The SIFT feature detector is a robust algorithm used to locate features, but cannot be extended directly to the 3D range point clouds since it requires dense pixel information, whereas the range voxels are sparsely distributed. This paper proposes an approach which enables SIFT application to locate scale and rotation invariant features in 3D point clouds. The algorithm then utilizes the known point correspondence registration algorithm in order to achieve real-time registration of 3D point clouds.