3D SUPER-RESOLUTION APPROACH FOR SPARSE LASER SCANNER DATA
Laser scanner point cloud has been emerging in Photogrammetry and computer vision to achieve high level tasks such as object tracking, object recognition and scene understanding. However, low cost laser scanners are noisy, sparse and prone to systematic errors. This paper proposes a novel 3D super r...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2015-08-01
|
| Series: | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | http://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/II-3-W5/151/2015/isprsannals-II-3-W5-151-2015.pdf |
| Summary: | Laser scanner point cloud has been emerging in Photogrammetry and computer vision to achieve high level tasks such as object tracking,
object recognition and scene understanding. However, low cost laser scanners are noisy, sparse and prone to systematic errors. This
paper proposes a novel 3D super resolution approach to reconstruct surface of the objects in the scene. This method works on sparse,
unorganized point clouds and has superior performance over other surface recovery approaches. Since the proposed approach uses
anisotropic diffusion equation, it does not deteriorate the object boundaries and it preserves topology of the object. |
|---|---|
| ISSN: | 2194-9042 2194-9050 |