Large scale visual odometry using stereo vision
Year: 2009
Type: conference-paper
Abstract: This paper presents a system for egomotion\nestimation using a stereo head camera. The\ncamera motion estimation is based on features\ntracked along a video sequence. The system\nalso estimates the tridimensional geometry of\nthe environment by fusing the visual information from multiple views. Furthermore, the paper presents comparisons between two different algorithms. The first one is by applying triangulation to 3D points. Motion estimation using 3D points suffers from the problem of nonisotropic\nnoise due to the large uncertainty in\ndepth estimation. To deal with this problem\nwe present results with a second approach that works directly in the disparity space. Experimental results using a mobile platform are presented. The experiments cover long distances in urban-like environments with the presence of dynamic objects. The system presented is part of a bigger project involving autonomous navigation using vision only.
Authors: A. Hernandez-Gutierrez, Juan Nieto, Teresa Vidal‐Calleja, E. Nebot
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W325323492