DocumentCode :
2795398
Title :
Quantitative Evaluation of Feature Extractors for Visual SLAM
Author :
Klippenstein, J. ; Zhang, Hong
Author_Institution :
Univ. of Alberta, Edmonton
fYear :
2007
fDate :
28-30 May 2007
Firstpage :
157
Lastpage :
164
Abstract :
We present a performance evaluation framework for visual feature extraction and matching in the visual simultaneous localization and mapping (SLAM) context. Although feature extraction is a crucial component, no qualitative study comparing different techniques from the visual SLAM perspective exists. We extend previous image pair evaluation methods to handle non-planar scenes and the multiple image sequence requirements of our application, and compare three popular feature extractors used in visual SLAM: the Harris corner detector, the Kanade-Lucas-Tomasi tracker (KLT), and the scale-invariant feature transform (SIFT). We present results from a typical indoor environment in the form of recall/precision curves, and also investigate the effect of increasing distance between image viewpoints on extractor performance. Our results show that all methods can be made to perform well, although it is possible to distinguish between the three. We conclude by presenting guidelines for selecting a feature extractor for visual SLAM based on our experiments.
Keywords :
SLAM (robots); edge detection; feature extraction; image matching; image sequences; mobile robots; optical tracking; robot vision; Harris corner detector; Kanade-Lucas-Tomasi tracker; mobile robot; multiple image sequence; performance evaluation framework; scale-invariant feature transform; visual SLAM; visual feature extraction; visual feature matching; visual simultaneous localization and mapping context; Cameras; Computer vision; Detectors; Eigenvalues and eigenfunctions; Feature extraction; Filters; Karhunen-Loeve transforms; Robot sensing systems; Robot vision systems; Simultaneous localization and mapping;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer and Robot Vision, 2007. CRV '07. Fourth Canadian Conference on
Conference_Location :
Montreal, Que.
Print_ISBN :
0-7695-2786-8
Type :
conf
DOI :
10.1109/CRV.2007.52
Filename :
4228535
Link To Document :
بازگشت