Related resources
Search for item elsewhere
University researcher(s)
Three-dimensional interest point detection and description using Speeded-Up Robust Features and histograms of oriented points
Salvador Pacheco-Gutierrez, Alexandru Stancu, Mohamed Mustafa, Eduard Codres, Bogdan Codres
In: System Theory, Control and Computing (ICSTCC), 2015 19th International Conference on: System Theory, Control and Computing (ICSTCC), 2015 19th International Conference on; 14 Oct 2015-16 Oct 2015; Cheile Gradistei, Romania. Romania: IEEE; 2015. p. 348-353.
Access to files
- FULL-TEXT.PDF (pdf)
Abstract
This article presents a novel systematic methodologyfor the detection of interest points in 3D point clouds and itscorresponding descriptors by using the information of an RGBcamera and a structured-light sensor. This is achieved by fusingSpeeded-Up Robust Features (SURF) in the image space, andhistograms that statistically represent the relationship of threedimensional geometric data around the interest points. The SURFalgorithm is implemented over an image whose pixel coordinateshave a direct corresponding 3D point, thus allowing the fusion ofboth approaches. By combining both methodologies, it is intentto define a set of interest points whose descriptors are able tomaintain the intrinsic characteristics of its constituent parts suchas repeatability, distinctiveness and robustness while remainingcompact and fast to compute. The detected points will be usefor both, localization and mapping of mobile robots in partiallyunknown environments.
Keyword(s)