Saved in:
Bibliographic Details
Main Authors: Koprucu, Nursena, Nigam, Meher Shashwat, Xu, Shicheng, Abere, Biruk, Dominici, Gabriele, Rodriguez, Andrew, Vadgama, Sharvaree, Inal, Berfin, Tono, Alberto
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.06693
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912046959820800
author Koprucu, Nursena
Nigam, Meher Shashwat
Xu, Shicheng
Abere, Biruk
Dominici, Gabriele
Rodriguez, Andrew
Vadgama, Sharvaree
Inal, Berfin
Tono, Alberto
author_facet Koprucu, Nursena
Nigam, Meher Shashwat
Xu, Shicheng
Abere, Biruk
Dominici, Gabriele
Rodriguez, Andrew
Vadgama, Sharvaree
Inal, Berfin
Tono, Alberto
contents Inspired by Geoffrey Hinton emphasis on generative modeling, To recognize shapes, first learn to generate them, we explore the use of 3D diffusion models for object classification. Leveraging the density estimates from these models, our approach, the Diffusion Classifier for 3D Objects (DC3DO), enables zero-shot classification of 3D shapes without additional training. On average, our method achieves a 12.5 percent improvement compared to its multiview counterparts, demonstrating superior multimodal reasoning over discriminative approaches. DC3DO employs a class-conditional diffusion model trained on ShapeNet, and we run inferences on point clouds of chairs and cars. This work highlights the potential of generative models in 3D object classification.
format Preprint
id arxiv_https___arxiv_org_abs_2408_06693
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DC3DO: Diffusion Classifier for 3D Objects
Koprucu, Nursena
Nigam, Meher Shashwat
Xu, Shicheng
Abere, Biruk
Dominici, Gabriele
Rodriguez, Andrew
Vadgama, Sharvaree
Inal, Berfin
Tono, Alberto
Computer Vision and Pattern Recognition
Artificial Intelligence
Computational Geometry
Inspired by Geoffrey Hinton emphasis on generative modeling, To recognize shapes, first learn to generate them, we explore the use of 3D diffusion models for object classification. Leveraging the density estimates from these models, our approach, the Diffusion Classifier for 3D Objects (DC3DO), enables zero-shot classification of 3D shapes without additional training. On average, our method achieves a 12.5 percent improvement compared to its multiview counterparts, demonstrating superior multimodal reasoning over discriminative approaches. DC3DO employs a class-conditional diffusion model trained on ShapeNet, and we run inferences on point clouds of chairs and cars. This work highlights the potential of generative models in 3D object classification.
title DC3DO: Diffusion Classifier for 3D Objects
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computational Geometry
url https://arxiv.org/abs/2408.06693