Please use this identifier to cite or link to this item: https://hdl.handle.net/10953/3304
Title: A GPU-Based Framework for Generating Implicit Datasets of Voxelized Polygonal Models for the Training of 3D Convolutional Neural Networks
Authors: Ogáyar-Anguita, Carlos-Javier
Rueda-Ruiz, Antonio-Jesús
Segura-Sánchez, Rafael-Jesús
Díaz-Medina, Miguel
García-Fernández, Ángel-Luis
Abstract: In this paper we present an efficient GPU-based framework to dynamically perform the voxelization of polygonal models for training 3D convolutional neural networks. It is designed to manage the dataset augmentation by using efficient geometric transformations and random vertex displacements in GPU. With the proposed system, every voxelization is carried out on-the-fly for directly feeding the network. The computing performance with this approach is much better than with the standard method, that carries out every voxelization of each model separately and has much higher data processing overhead. The core voxelization algorithm works by using the standard rendering pipeline of the GPU. It generates binary voxels for both the interior and the surface of the models. Moreover, it is simple, concise and very compatible with commodity hardware, as it neither uses complex data structures nor needs vendor-specific hardware or additional dependencies. This framework dramatically reduces the input/output operations, and completely eliminates the storage footprint of the voxelization dataset, managing it as an implicit dataset.
Keywords: Voxelization
B-Rep
Boundary representation
Polygonal meshes
Convolutional neural network
3D-CNN
Geometric deep learning
Issue Date: 10-Jan-2020
metadata.dc.description.sponsorship: Ministerio de Ciencia, Innovación y Universidades. Proyecto RTI2018-099638-B-I00
Publisher: IEEE
Citation: C. J. Ogayar-Anguita, A. J. Rueda-Ruiz, R. J. Segura-Sánchez, M. Díaz-Medina and Á. L. García-Fernández, "A GPU-Based Framework for Generating Implicit Datasets of Voxelized Polygonal Models for the Training of 3D Convolutional Neural Networks," in IEEE Access, vol. 8, pp. 12675-12687, 2020, doi: 10.1109/ACCESS.2020.2965624.
Appears in Collections:DI-Artículos

Files in This Item:
File Description SizeFormat 
paper.pdfAccepted version29,6 MBAdobe PDFView/Open


This item is protected by original copyright