Learning Representations and Generative Models for 3D Point Clouds

by   Panos Achlioptas, et al.

Three-dimensional geometric data offer an excellent domain for studying representation learning and generative modeling. In this paper, we look at geometric data represented as point clouds. We introduce a deep autoencoder (AE) network with state-of-the-art reconstruction quality and generalization ability. The learned representations outperform existing methods for 3D recognition tasks and enable basic shape editing via simple algebraic manipulations, such as semantic part editing, shape analogies and shape interpolation. We perform a thorough study of different generative models including: GANs operating on the raw point clouds, significantly improved GANs trained in the fixed latent space our AEs and Gaussian mixture models (GMM). For our quantitative evaluation we propose measures of sample fidelity and diversity based on matchings between sets of point clouds. Interestingly, our careful evaluation of generalization, fidelity and diversity reveals that GMMs trained in the latent space of our AEs produce the best results.


page 5

page 9

page 10

page 19

page 20


Generative Models for 3D Point Clouds

Point clouds are rich geometric data structures, where their three dimen...

Latent-Space Laplacian Pyramids for Adversarial Representation Learning with 3D Point Clouds

Constructing high-quality generative models for 3D shapes is a fundament...

Adversarial Autoencoders for Generating 3D Point Clouds

Deep generative architectures provide a way to model not only images, bu...

Geometric Disentanglement for Generative Latent Shape Models

Representing 3D shape is a fundamental problem in artificial intelligenc...

Learning Generative Models of Shape Handles

We present a generative model to synthesize 3D shapes as sets of handles...

Escape from Cells: Deep Kd-Networks for the Recognition of 3D Point Cloud Models

We present a new deep learning architecture (called Kd-network) that is ...

Learning Smooth Neural Functions via Lipschitz Regularization

Neural implicit fields have recently emerged as a useful representation ...

Code Repositories


Auto-encoding & Generating 3D Point-Clouds.

view repo

Please sign up or login with your details

Forgot password? Click here to reset