Instance Segmentation of Point Clouds using Deep Learning

The main objective of this thesis is to implement a deep network architecture to segment and instantiate objects on 3d point clouds using pointnet, a novel 3d deep learning network that segments 3d point clouds in semantically meaningful classes, as a baseline.

Detalhes bibliográficos
Autor: Pérez Layedra, Gerardo Francisco
Tipo de documento: dissertação
Data de publicação:2018
País:España
Recursos:Universitat Politècnica de Catalunya (UPC)
Repositório:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglês
OAI Identifier:oai:upcommons.upc.edu:2117/117737
Acesso em linha:https://hdl.handle.net/2117/117737
Access Level:Acceso aberto
Palavra-chave:Machine learning
Cloud computing
point cloud
instantiation
segmentation
pointnet
deep learning
tensorflow
python
Aprenentatge automàtic
Computació en núvol
Àrees temàtiques de la UPC::Informàtica
Descrição
Resumo:The main objective of this thesis is to implement a deep network architecture to segment and instantiate objects on 3d point clouds using pointnet, a novel 3d deep learning network that segments 3d point clouds in semantically meaningful classes, as a baseline.