Quantitative and flexible 3D shape dataset augmentation via latent space embedding and deformation learning

DOI10.1016/J.CAGD.2019.04.017MaRDI QIDQ2010274zbMATH OpenOpenAlexWikidataFDO

Authors Jiarui Liu, Qing Xia, Shuai Li, Aimin Hao, Hong Qin

Publication date 27 November 2019

Published in Computer Aided Geometric Design (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.cagd.2019.04.017

zbMATH Keywords

data augmentationpoint cloudsGaussian process latent-variable model

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Computer graphics; computational geometry (digital and algorithmic aspects) (68U05) Machine vision and scene understanding (68T45)

Recommendations

• Deep neural networks for geometric shape deformation
• Learning geometry-aware joint latent space for simultaneous multimodal shape generation
• Inferring 3D shapes from image collections using adversarial networks
• Unsupervised 3D shape segmentation and co-segmentation via deep learning
• Shape part transfer via semantic latent space factorization

Cites work

• Learning a manifold of fonts
• Poisson shape interpolation
• Principal Curves
• Probabilistic non-linear principal component analysis with Gaussian process latent variable models
• Real-time nonlinear shape interpolation
• The earth mover's distance as a metric for image retrieval

Cited in

(8)

• An inverse scattering approach for geometric body generation: a machine learning perspective
• Inferring 3D shapes from image collections using adversarial networks
• Learning geometry-aware joint latent space for simultaneous multimodal shape generation
• Deep neural networks for geometric shape deformation
• TPNet: a novel mesh analysis method via topology preservation and perception enhancement
• Fast mesh data augmentation via Chebyshev polynomial of spectral filtering
• Structure-aware shape correspondence network for 3D shape synthesis
• Shape part transfer via semantic latent space factorization

Describes a project that uses

Uses Software

• PointNet
• EMD

This page was built for publication: Quantitative and flexible 3D shape dataset augmentation via latent space embedding and deformation learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2010274)