Automatic object extraction from aerial imagery—a survey focusing on buildings
Computer Vision and Image Understanding
Using 3D and Ancillary Media to Train Construction Workers
IEEE MultiMedia
Universal Traversal Sequences with Backtracking
CCC '01 Proceedings of the 16th Annual Conference on Computational Complexity
Hi-index | 0.00 |
It is a common requirement to build a 2D map of a 3D virtual scene for path planning, but it is difficult or tedious. So we propose a multiple-level grid algorithm to solve the problem. The original idea lies in the new method on distinguishing a grid unit's accessibility. We argue for an approach by employing two navigating lines, which are right-angle intersection and stand for the unit's accessibility in four directions. Moreover, another advantage of it is to dynamically divide the virtual scene into different sizes of grids at different precisions. With these multi-precision grids, we can approach objects at any granularity. Lastly, experiments are performed to test the effectiveness of the algorithm.