The method consists of two core components: an efficient selective reconstruction algorithm, based on geometric convection, that simplifies the input point set while reconstructing a surface, and a local update algorithm that dynamically refines or coarsens the reconstructed surface according to specific local sampling constraints.
As there is no information about the orientation of the beam, all A-scans were oriented vertically, reconstructing a surface that appears erroneously flat.
Instead of determining feature correspondences, these methods aim to recover the camera pose directly from the image data, by reconstructing a surface-based depth map for the image.
An adaptive technique to reconstruct a smooth surface from Bézier patches is presented in this paper.
To reconstruct an object surface from a set of surface points promptly, it is a necessity to reduce the data points.
The SEM software (Phenom Pro Suite, Phenom-World BV, The Netherlands) was used to quantity the surface morphology by reconstructing a 3D surface, from which mean surface roughness values (Ra) and mean surface depth values (Rz) were calculated.
This paper presents a robust surface fitting algorithm to reconstruct a high fidelity surface from measured discrete points while the surface smoothness can be ensured as well.
Despite using 32 cameras, not all are required; Oliver Bao, head of Depth Analysis R&D, said that the extra samples allow the team to "reconstruct a better 3D surface in general" and to "remove reconstruction noise a little more easily".
In this paper, we propose a nonlinear PDE model for reconstructing a regular surface from sampled data.
"With the rover we take images every day and we can reconstruct a 3D map of the surface," says Calef. "With the navigation cameras we get down to a resolution of 3cm and with the other cameras and instruments on board we can resolve that to the size of sand grains".
