Generative and Discriminative Voxel Modeling with Convolutional Neural Networks

Andrew Brock, Theodore Lim, J. M. Ritchie, Nick Weston
8/15/2016
cs.CVcs.HCcs.LG

Official code: ajbrock/Generative-and-Discriminative-Voxel-Modeling (214 stars, MIT, last commit 2016-09-11).

Abstract

When working with three-dimensional data, choice of representation is key. We explore voxel-based models, and present evidence for the viability of voxellated representations in applications including shape modeling and object classification. Our key contributions are methods for training voxel-based variational autoencoders, a user interface for exploring the latent space learned by the autoencoder, and a deep convolutional neural network architecture for object classification. We address challenges unique to voxel-based representations, and empirically evaluate our models on the ModelNet benchmark, where we demonstrate a 51.5% relative improvement in the state of the art for object classification.

View on arXivView PDF

Code Implementations(1)

ajbrock/Generative-and-Discriminative-Voxel-ModelingOfficial100%
MIT

Voxel-Based Variational Autoencoders, VAE GUI, and Convnets for Classification

21461TeX, MATLABJul 30, 201610 years agoMIT
convnetdeep-learningguimachine-learningvariational-autoencoders

Similar Papers

Native and Compact Structured Latents for 3D Generation

83% match4 repos8.8k2025

SnakeVoxFormer: Transformer-based Single Image\\Voxel Reconstruction with Run Length Encoding

82% match2023

ADASSM: Adversarial Data Augmentation in Statistical Shape Models From Images

82% match2023

VortSDF: 3D Modeling with Centroidal Voronoi Tesselation on Signed Distance Field

81% match2024

GVSynergy-Det: Synergistic Gaussian-Voxel Representations for Multi-View 3D Object Detection

81% match2025

VoMP: Predicting Volumetric Mechanical Property Fields

81% match2025

Cite this paper

@article{brock2016generative,
  title  = {Generative and Discriminative Voxel Modeling with Convolutional Neural Networks},
  author = {Andrew Brock and Theodore Lim and J. M. Ritchie and Nick Weston},
  year   = {2016},
  eprint = {1608.04236},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1608.04236}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1608.04236.svg)](https://codeofpaper.com/paper/1608.04236)

Discussion