Hough transform to extract 3D information from images of different viewpoints
We propose a method to extract 3D information by projecting back the feature points in the images into the 3D space. We divide the 3D space into voxels and apply the Hough transform technique by giving a vote to every voxel on the backprojected lines. However, a simple voting rule raises some problems. We propose a wellcontrived voting and evaluation rule to solve these problems, which we call the double backprojection method (DBP). The octree representation of objects can be adopted to DBP to allow multi-resolution analysis and to increase calculation efficiency. Experimental results are also described.
Unable to display preview. Download preview PDF.
- 1.T.Hamano, T.Yasuno and K.Ishii: Direct Estimation of Structure from Non-linear Motion by Voting Algorithm without Tracking, Proc. of ICPR'92, vol.1, pp.505–508, 1992.Google Scholar
- 2.S.Kawato: 3D shape recovery by octree voting technique, Proc. of SPIE, vol.1820, pp.40–49, 1992.Google Scholar
- 3.M. Potmesil: Generating octree models of 3D objects from their silhouettes in a sequence of images, Comput. Vision, Graphics & Image Process., vol.40, no.1, pp.1–29, 1987.Google Scholar
- 4.H.Samet: The Design and Analysis of Spatial Data Structures, pp.493, Addison Wesley, 1989.Google Scholar
- 5.S.Srivastava and N.Ahuja: Octree generation from object silhouettes in perspective views, Comput. Vision, Graphics & Image Process., vol.49, pp.68–84, 1990.Google Scholar
- 6.R.Szeliski: Real-time octree generation from rotating objects, Tech.Rep. 90/12, DEC Cambridge Research Lab., p.30, Dec. 1990.Google Scholar