HIERARCHICAL OBB-SPHERE TREE FOR LARGE-SCALE RANGE DATA MANAGEMENT
- HIERARCHICAL OBB-SPHERE TREE FOR LARGE-SCALE RANGE DATA MANAGEMENT
- Nguyen Thuy Hoang Phong; 홍승표; 김진욱
- Range data management; hierarchical
bounding volume; range data association; 3D scene reconstruction
- Issue Date
- International Conference on Image Processing
- Range data association in 3D scene reconstruction often runs pair-wise scan matching for all pairs of input data. However, the computationally expensive matching operation is
false and wasteful for many irrelevant image pairs. OBBsphere tree presented in this paper accelerates the process by choosing intelligently a subset of related-pair candidates. The proposed data structure is a sphere tree with every leaf node containing a 3D point set converted from a depth image and bounded by one sphere and one tighter-fitting shape, oriented bounding box (OBB). This approach exploits the simplicity of spheres in hierarchical tree construction and intersection
test to reject objects far apart as well as the compactness of OBBs in the intersection test refinement. Number of possible pairs was cut down more than half for some real datasets in experiment. OBB-sphere tree is able to deal with available range data or on-the-fly data insertion. It can also handle the loop closure detection and support the rendering via view frustum culling and ray tracing test.
- Appears in Collections:
- KIST Publication > Conference Paper
- Files in This Item:
There are no files associated with this item.
- RIS (EndNote)
- XLS (Excel)
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.