Outdoor place recognition in urban environments using straight lines
Authors:
- 1 May 2014
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10504729,p. 5550-5557
Abstract
In this paper, we propose a visual place recognition algorithm which uses only straight line features in challenging outdoor environments. Compared to point features used in most existing place recognition methods, line features are easily found in man-made environments and more robust to environmental changes such as illumination, viewing direction, or occlusion because they are more likely to be extracted from structures. Candidate matches are found using a vocabulary tree and their geometric consistency is verified by a motion estimation algorithm using line segments. The proposed algorithm operates in real-time, and it is tested with a challenging real-world dataset with more than 10,000 database images acquired in urban driving scenarios.Keywords
This publication has 17 references indexed in Scilit:
- A vertical and floor line-based monocular SLAM system for corridor environmentsInternational Journal of Control, Automation, and Systems, 2012
- Moving in stereo: Efficient structure and motion using lines2009
- MSLD: A robust descriptor for line matchingPattern Recognition, 2009
- A visual bag of words method for interactive qualitative localization and mapping2007
- Scalable Recognition with a Vocabulary Tree2006
- Wide-Baseline Stereo Matching with Line Segments2005
- An efficient solution to the five-point relative pose problemIEEE Transactions on Pattern Analysis and Machine Intelligence, 2004
- Video Google: a text retrieval approach to object matching in videos2003
- In defense of the eight-point algorithmIEEE Transactions on Pattern Analysis and Machine Intelligence, 1997
- Estimating motion and structure from correspondences of line segments between two perspective imagesIEEE Transactions on Pattern Analysis and Machine Intelligence, 1995