Abstract
We propose a new method of creating a complete model of a curved object from multiple range images acquired by showing it at different poses. The pose of the object is changed by a manipulator in order to view the object from some specified viewpoints. The pose is planned after each new image is merged into a unified representation. A rating function for the planning is defined to take into consideration the factors such as possibility of merging new data, registration accuracy and control point selection.
Original language | English |
---|---|
Pages | 68-75 |
Number of pages | 8 |
Publication status | Published - 1997 |
Externally published | Yes |
Event | Proceedings of the 1997 1st International Conference on Recent Advances in 3-D Digital Imaging and Modeling - Ottawa, Can Duration: May 12 1997 → May 15 1997 |
Other
Other | Proceedings of the 1997 1st International Conference on Recent Advances in 3-D Digital Imaging and Modeling |
---|---|
City | Ottawa, Can |
Period | 5/12/97 → 5/15/97 |
All Science Journal Classification (ASJC) codes
- Computer Vision and Pattern Recognition