The accuracy of the HRBF networks

Author

Ferrari, S. ; Frosio, I. ; Piuri, V. ; Borghese, N.A.

Author_Institution

Dept. of Inf. Technol., Milano Univ., Crema, Italy

Volume

1

fYear

2004

Firstpage

482

Abstract

A procedure for the construction of 3D surfaces from range data in real-time is here described and discussed. The main goal is to observe the accuracy of the reconstruction through the phases of the 3D model reconstruction. The process is based on the connectionist model, named hierarchical radial basis functions networks (HRBF), which has been proved effective in the reconstruction of smooth surfaces from sparse noisy points. The network goal is to achieve a uniform reconstruction error, equal to measurement error, by stacking non-complete grids of Gaussians at decreasing scales. The HRBF properties allow reconstructing meshes from range data in real-time.

Keywords

Gaussian processes; image reconstruction; measurement errors; mesh generation; radial basis function networks; real-time systems; 3D model reconstruction; 3D scanner; 3D surface construction; connectionist model; hierarchical radial basis functions networks; measurement error; meshes reconstruction; multiscale surface; noise filtering; non-complete grids stacking; real-time meshing; smooth surfaces; sparse noisy points; uniform reconstruction error; Computer science; Feedback; Gaussian processes; Humans; Information technology; Merging; Mesh generation; Sampling methods; Stacking; Surface reconstruction;

fLanguage

English

Publisher

ieee

Conference_Titel

Instrumentation and Measurement Technology Conference, 2004. IMTC 04. Proceedings of the 21st IEEE

ISSN

1091-5281

Print_ISBN

0-7803-8248-X

Type

conf

DOI

10.1109/IMTC.2004.1351093

Filename

1351093