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

  1. Orts Escolano, Sergio
Dirigida por:
  1. José García Rodríguez Director
  2. Miguel Cazorla Quevedo Codirector

Universidad de defensa: Universitat d'Alacant / Universidad de Alicante

Fecha de defensa: 15 de enero de 2014

Tribunal:
  1. Manuel Graña Romay Presidente/a
  2. Juan Manuel García Chamizo Secretario
  3. Alexandra Psarrou Vocal
Departamento:
  1. TECNOLOGIA INFORMATICA Y COMPUTACION

Tipo: Tesis

Teseo: 355697 DIALNET lock_openRUA editor

Resumen

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 thesis proposed the use of Self-Organizing Maps (SOMs) as a 3D representation model. In particular, we proposed the use of the Growing Neural Gas (GNG) network, which has been successfully used for clustering, pattern recognition and topology representation of multi-dimensional data. Until now, Self-Organizing Maps have been primarily computed offline and their application in 3D data has mainly focused on free noise models, without considering time constraints. It is proposed a hardware implementation leveraging the computing power of modern GPUs, which takes advantage of a new paradigm coined as General-Purpose Computing on Graphics Processing Units (GPGPU). The proposed methods were applied to different problems and applications in the area of computer vision such as the recognition and localization of objects, visual surveillance or 3D reconstruction.