Visual odometry research based on efficient feature tracking
Ding Lianghong, Tao Li, Wang Runxiao · 2012
Visual odometry can use feature tracking to estimate the motion of robot in sequence stereo frames. High quality feature point set can be obtained by all kinds of detection and removal methods through the each process of visual odometry. Several methods have been summarized in the paper and applied in our experiments. The result shows that these methods are better than single RANSAC algorithm.