• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.55.2024.tde-02072024-085526
Document
Author
Full name
Guilherme Hideo Tubone
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Carlos, 2024
Supervisor
Committee
Batista Neto, João do Espírito Santo (President)
Casaca, Wallace Correa de Oliveira
Gois, João Paulo
Paiva Neto, Afonso
Title in English
3D Facial Reconstruction from Point Clouds with Hermite Radial Basis Functions and Multilevel Partition of Unity
Keywords in English
Faces
Hermite radial basis functions
Multilevel partition of unity
Point clouds
Surface reconstruction
Abstract in English
Point cloud surface reconstruction is the process of transforming raw data from 3D scanners into a digital representation of an object surface. This process has applications in various fields, and as a result, many methods have been proposed to solve this problem over the years. In this work, we focus on methods that generate surfaces represented as level sets of implicit functions. Specifically, we propose a method based on Hermite Radial Basis Functions (HRBF) with Multilevel Partition of Unity (MPU). The HRBF allows a precise and detailed representation of the surface geometry, even when dealing with nonuniform data, while the MPU enables a efficient and adaptive reconstruction process. To demonstrate the efficacy of our method, we applied it to a dataset of human faces. The results show that the method can successfully reconstruct the facial geometry, capturing details of the surfaces.
Title in Portuguese
Reconstrução Facial 3D a Partir de Nuvem de Pontos com Funções de Base Radial de Hermite e Partição da Unidade Multinível
Keywords in Portuguese
Faces
Funções de base radial de Hermite
Nuvem de pontos
Partição da unidade multinível
Reconstrução de superfícies
Abstract in Portuguese
A reconstrução de superfície de nuvem de pontos é o processo de transformar dados brutos de scanners 3D em uma representação digital da superfície de um objeto. Esse processo tem aplicações em vários campos, e por isso, muitos métodos têm sido propostos ao longo dos anos para resolver esse problema. Neste trabalho, nos concentramos em métodos que geram superfícies representadas como níveis de funções implícitas. Especificamente, propomos um método baseado em Funções de Base Radial de Hermite (HRBF) com Partição de Unidade Multinível (MPU). A HRBF permite uma representação precisa e detalhada da geometria da superfície, mesmo quando se trabalha com dados não uniformes, enquanto o MPU permite um processo de reconstrução eficiente e adaptativo. Para demonstrar a eficácia do nosso método, aplicamos-no a um conjunto de dados de faces humanas. Os resultados mostram que o método pode reconstruir com sucesso a geometria facial, capturando detalhes das superfícies.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2024-07-02
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.