Generation and weighting of 3D point correspondences for improved registration of RGB-D data
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 sit...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2013-10-01
|
| Series: | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/II-5-W2/127/2013/isprsannals-II-5-W2-127-2013.pdf |
| _version_ | 1819012548940267520 |
|---|---|
| author | K. Khoshelham D. R. Dos Santos G. Vosselman |
| author_facet | K. Khoshelham D. R. Dos Santos G. Vosselman |
| author_sort | K. Khoshelham |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-12-21T01:45:49Z |
| format | Article |
| id | doaj.art-ca8379b7a59643cb94186bc5f5bcd0ae |
| institution | Directory Open Access Journal |
| issn | 2194-9042 2194-9050 |
| language | English |
| last_indexed | 2024-12-21T01:45:49Z |
| publishDate | 2013-10-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| spelling | doaj.art-ca8379b7a59643cb94186bc5f5bcd0ae2022-12-21T19:20:02ZengCopernicus PublicationsISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences2194-90422194-90502013-10-01II-5-W212713210.5194/isprsannals-II-5-W2-127-2013Generation and weighting of 3D point correspondences for improved registration of RGB-D dataK. Khoshelham0D. R. Dos Santos1G. Vosselman2Faculty of Geo-Information Science and Earth Observation, University of Twente, the NetherlandsFederal University of Parana, Curitiba, BrazilFaculty of Geo-Information Science and Earth Observation, University of Twente, the NetherlandsRegistration 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.https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/II-5-W2/127/2013/isprsannals-II-5-W2-127-2013.pdf |
| spellingShingle | K. Khoshelham D. R. Dos Santos G. Vosselman Generation and weighting of 3D point correspondences for improved registration of RGB-D data ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| title | Generation and weighting of 3D point correspondences for improved registration of RGB-D data |
| title_full | Generation and weighting of 3D point correspondences for improved registration of RGB-D data |
| title_fullStr | Generation and weighting of 3D point correspondences for improved registration of RGB-D data |
| title_full_unstemmed | Generation and weighting of 3D point correspondences for improved registration of RGB-D data |
| title_short | Generation and weighting of 3D point correspondences for improved registration of RGB-D data |
| title_sort | generation and weighting of 3d point correspondences for improved registration of rgb d data |
| url | https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/II-5-W2/127/2013/isprsannals-II-5-W2-127-2013.pdf |
| work_keys_str_mv | AT kkhoshelham generationandweightingof3dpointcorrespondencesforimprovedregistrationofrgbddata AT drdossantos generationandweightingof3dpointcorrespondencesforimprovedregistrationofrgbddata AT gvosselman generationandweightingof3dpointcorrespondencesforimprovedregistrationofrgbddata |