Some people are mistakenly trying to apply image processing techniques to point cloud processing, e.g., edge detection in point cloud. Edges by image processing are boundary of 2-D object regions. And, with point cloud processing? Point clouds are already boundary of 3-D object volumes. Edge detection in point cloud is equivalent to determining the marginal end-points of an edge in image. Both the marginal end-points of an edge in image and the edges in point cloud are highly unstable and unreliable. As we can determine accurately a 2-D point by intersecting 2 curves fitted to the edges in image, we can determine accurately a 3-D edge curve by intersecting 2 surfaces fitted to the point cloud. In summary, Correct data processing would be: Curve fitting to edges in image. Surface fitting to point cloud. Ineffective data processing would be: Determining the marginal end-points of an edge in image. Edge detection in point cloud. Plane, box, cuboid, sphere, cylinder, cone, torus, object recognition.
Some Thoughts on 2-D/3-D Information Processing - YouTube |
| Likes | Dislikes |
| 608 views views | 377 followers |
| Science & Technology Creative Commons Attribution licence (reuse allowed) | Upload TimePublished on 7 Nov 2015 |
3d printer,3ds,3d ultrasound,3ds emulator,3d printing,3ds games,3ds xl,3d tattoos,3d pen,3d warehouse,3d pictures,3d tuning,3d nails,3d doritos,3ds max,3d shapes,3d printed house,3d scanner,3d printing service,3dmark
Không có nhận xét nào:
Đăng nhận xét