A Solution to the Next Best View Problem for Automated Surface Acquisition
IEEE Transactions on Pattern Analysis and Machine Intelligence
Active modeling of 3-D objects: planning on the next best pose (NBP) for acquiring range images
NRC '97 Proceedings of the International Conference on Recent Advances in 3-D Digital Imaging and Modeling
Computations on a spherical view space for efficient planning of viewpoints in 3-D object modeling
3DIM'99 Proceedings of the 2nd international conference on 3-D digital imaging and modeling
Automatic sensor placement for model-based robot vision
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Hi-index | 0.00 |
For automatic modeling of three-dimensional unknown objects or environments by an active robot system, the perception strategy is a critical issue since sensing operations have to be decided dynamically and sequentially. This paper presents a tree-structured approach for merging, splitting, and viewpoint deciding. Further development of the algorithms will lead to a very robust method for applying in such complex vision tasks. Repetitive processes, system states, and complete condition are also discussed for exploring the unknown portion of the object or the environment. This paper also treats uncertainties and occlusions, as well as many robot environmental or sensing constraints. Redundant viewpoints are eliminated and new views are planned according to a max-min criterion for achieving lowest cost for robot operations.