Homography Based Kalman Filter for Mosaic Building. Applications to UAV position estimation
• Probabilistic framework where uncertainties can be considered in the mosaic building process.
• Inter-image relations are modeled by homographies and its covariances.
• When a loop is closed, the estimation errors can be compensated and propagated to the rest of the mosaic.
• The relations among images in the mosaic can be used to localize the robot.
F. Caballero1, L. Merino2, J Ferruz1 and A. Ollero1 1 University of Seville, Seville (Spain)
2 Pablo de Olavide University, Seville (Spain)
Parking at LAAS, France.Mosaic composed by 450 images