Reduced egomotion estimation drift using omnidirectional views
This work addresses drift reduction in egomotion estimation for applications like robotics or autonomous navigation, but it is incremental as it builds on existing camera-based methods by adding an omnidirectional view.
The paper tackles the problem of increasing drift in camera motion estimation over long sequences by proposing a method that uses an omnidirectional camera alongside a perspective camera to exploit correspondences between their images, resulting in improved estimation accuracy as shown in simulated and real experiments.
Estimation of camera motion from a given image sequence becomes degraded as the length of the sequence increases. In this letter, this phenomenon is demonstrated and an approach to increase the estimation accuracy is proposed. The proposed method uses an omnidirectional camera in addition to the perspective one and takes advantage of its enlarged view by exploiting the correspondences between the omnidirectional and perspective images. Simulated and real image experiments show that the proposed approach improves the estimation accuracy.