Saved in:
Bibliographic Details
Main Authors: Fan, Jiajie, Trigui, Amal, Bonfanti, Andrea, Dietrich, Felix, Bäck, Thomas, Wang, Hao
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.06485
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913725736288256
author Fan, Jiajie
Trigui, Amal
Bonfanti, Andrea
Dietrich, Felix
Bäck, Thomas
Wang, Hao
author_facet Fan, Jiajie
Trigui, Amal
Bonfanti, Andrea
Dietrich, Felix
Bäck, Thomas
Wang, Hao
contents Recent advancements in learning latent codes derived from high-dimensional shapes have demonstrated impressive outcomes in 3D generative modeling. Traditionally, these approaches employ a trained autoencoder to acquire a continuous implicit representation of source shapes, which can be computationally expensive. This paper introduces a novel framework, spectral-domain diffusion for high-quality shape generation SpoDify, that utilizes singular value decomposition (SVD) for shape encoding. The resulting eigenvectors can be stored for subsequent decoding, while generative modeling is performed on the eigenfeatures. This approach efficiently encodes complex meshes into continuous implicit representations, such as encoding a 15k-vertex mesh to a 512-dimensional latent code without learning. Our method exhibits significant advantages in scenarios with limited samples or GPU resources. In mesh generation tasks, our approach produces high-quality shapes that are comparable to state-of-the-art methods.
format Preprint
id arxiv_https___arxiv_org_abs_2503_06485
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Mesh Is Worth 512 Numbers: Spectral-domain Diffusion Modeling for High-dimension Shape Generation
Fan, Jiajie
Trigui, Amal
Bonfanti, Andrea
Dietrich, Felix
Bäck, Thomas
Wang, Hao
Computer Vision and Pattern Recognition
Recent advancements in learning latent codes derived from high-dimensional shapes have demonstrated impressive outcomes in 3D generative modeling. Traditionally, these approaches employ a trained autoencoder to acquire a continuous implicit representation of source shapes, which can be computationally expensive. This paper introduces a novel framework, spectral-domain diffusion for high-quality shape generation SpoDify, that utilizes singular value decomposition (SVD) for shape encoding. The resulting eigenvectors can be stored for subsequent decoding, while generative modeling is performed on the eigenfeatures. This approach efficiently encodes complex meshes into continuous implicit representations, such as encoding a 15k-vertex mesh to a 512-dimensional latent code without learning. Our method exhibits significant advantages in scenarios with limited samples or GPU resources. In mesh generation tasks, our approach produces high-quality shapes that are comparable to state-of-the-art methods.
title A Mesh Is Worth 512 Numbers: Spectral-domain Diffusion Modeling for High-dimension Shape Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.06485