Generation and weighting of 3D point correspondences for improved registration of RGB-D data
Other literature type
Dos Santos, D. R.
Registration of RGB-D data using visual features is often influenced by errors in the transformation of visual features to 3D space as
well as the random error of individual 3D points. In a long sequence, these errors accumulate and lead to inaccurate and deformed
point clouds, particularly in situations where loop closing is not feasible. We present an epipolar search method for accurate
transformation of the keypoints from 2D to 3D space, and define weights for the 3D points based on the theoretical random error of
depth measurements. Our results show that the epipolar search method results in more accurate 3D correspondences. We also
demonstrate that weighting the 3D points improves the accuracy of sensor pose estimates along the trajectory.