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 improve...
This lecture presents: Session two- the integration of cycling and public transport; Objectives of this workshop; Rationale for integration of bike and PT; Density of pt stations: centre of Rio de Janeiro; Finding the optimal mix of modes; Buses and bikes: friends or foes?;Public transport integration.
