Learning visual landmarks for pose estimation
- 20 January 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 3 (10504729) , 1972-1978
- https://doi.org/10.1109/robot.1999.770397
Abstract
We present an approach to vision-based mobile robot localization, even without an a-priori pose estimate. This is accomplished by learning a set of visual features called image-domain landmarks. The landmark learning mechanism is designed to be applicable to a wide range of environments. Each landmark is detected as a focal extremum of a measure of uniqueness and represented by an appearance-based encoding. Localization is performed using a method that matches observed landmarks to learned prototypes and generates independent position estimates for each match. The independent estimates are then combined to obtain a final position estimate, with an associated uncertainty. Quantitative experimental evidence is presented that demonstrates that accurate pose estimates can be obtained, despite changes to the environment.Keywords
This publication has 13 references indexed in Scilit:
- Robot localization from landmarks using recursive total least squaresPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Vision-based robot localization without explicit object modelsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Combinatorial optimization applied to variable scale 2D model matchingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Robotic sightseeing-a method for automatically creating virtual environmentsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- EigenTracking: Robust matching and tracking of articulated objects using a view-based representationPublished by Springer Nature ,1996
- View-based and modular eigenspaces for face recognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1994
- Mobile robot localization by tracking geometric beaconsIEEE Transactions on Robotics and Automation, 1991
- Locating a robot with angle measurementsJournal of Symbolic Computation, 1990
- On the Representation and Estimation of Spatial UncertaintyThe International Journal of Robotics Research, 1986
- Convergence rates of "thin plate" smoothing splines wihen the data are noisyPublished by Springer Nature ,1979