A three-dimensional representation method for noisy point clouds based on growing self-organizing maps accelerated on GPUs

A three-dimensional representation method for noisy point clouds based on growing self-organizing maps accelerated on GPUs

The research described in this thesis was motivated by the need of a robust model capable of representing 3D data obtained with 3D sensors, which are inherently noisy. In addition, time constraints have to be considered as these sensors are capable of providing a 3D data stream in real time. This th...

Full description

Bibliographic Details
Main Author: Orts-Escolano, Sergio
Other Authors: Garcia-Rodriguez, Jose
Format: Doctoral Thesis
Language:English
Published: Universidad de Alicante 2014
Subjects:
3D representation method
Growing neural gas
Self-organizing maps
Topology preservation
Parallel computing
CUDA
Real-time
Point cloud
3D reconstruction
GPGPU
RGBD
Noisy 3D data
Object recognition
Arquitectura y Tecnología de Computadores
Online Access:http://hdl.handle.net/10045/36484

Internet

http://hdl.handle.net/10045/36484

Similar Items