We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Graphics

Title: Learning to Generate 3D Shapes from a Single Example

Abstract: Existing generative models for 3D shapes are typically trained on a large 3D dataset, often of a specific object category. In this paper, we investigate the deep generative model that learns from only a single reference 3D shape. Specifically, we present a multi-scale GAN-based model designed to capture the input shape's geometric features across a range of spatial scales. To avoid large memory and computational cost induced by operating on the 3D volume, we build our generator atop the tri-plane hybrid representation, which requires only 2D convolutions. We train our generative model on a voxel pyramid of the reference shape, without the need of any external supervision or manual annotation. Once trained, our model can generate diverse and high-quality 3D shapes possibly of different sizes and aspect ratios. The resulting shapes present variations across different scales, and at the same time retain the global structure of the reference shape. Through extensive evaluation, both qualitative and quantitative, we demonstrate that our model can generate 3D shapes of various types.
Comments: SIGGRAPH Asia 2022; 19 pages (including 6 pages appendix), 17 figures. Project page: this http URL
Subjects: Graphics (cs.GR); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
DOI: 10.1145/3550454.3555480
Cite as: arXiv:2208.02946 [cs.GR]
  (or arXiv:2208.02946v2 [cs.GR] for this version)

Submission history

From: Rundi Wu [view email]
[v1] Fri, 5 Aug 2022 01:05:32 GMT (5336kb,D)
[v2] Thu, 22 Sep 2022 01:06:48 GMT (5337kb,D)

Link back to: arXiv, form interface, contact.