Presentation

<
MeshFlow: Mesh Generation with Equivariant Flow Matching
Session3D Generation
DescriptionMeshFlow is a powerful diffusion-based framework utilizing equivariant modeling to transform random triangle soups into high-quality meshes in less than one second.
Authors
Qi Sun
City University of Hong Kong
Kiyohiro Nakayama
Stanford University
Jing Yan
Cornell Tech
Qixing Huang
The University of Texas at Austin
Alexander Rush
Cornell Tech
Leonidas Guibas
Stanford University
Gordon Wetzstein
Stanford University
Jing Liao
City University of Hong Kong
Guandao Yang
The University of Texas at Austin
Event Type
Technical Paper
TimeTuesday, 21 July 202611:35am - 11:45am PDT
LocationRoom 403 B
Digital Library PDF
Session Time & Location
Sunday, 19 July 20266:00pm - 8:45pm PDTHall K
Tuesday, 21 July 202610:45am - 12:35pm PDTRoom 403 B
Keywords
Artificial Intelligence/Machine Learning
Modeling
Registration Categories
Full Conference Supporter
Full Conference
Next PresentationNext PresentationStrips as Tokens: Artist Mesh Generation with Native UV Segmentation
Similar Presentations
Matérn Noise for Triangulation-Agnostic Flow Matching on Meshes
Fast and Exact Winding Numbers for Triangle Meshes
Nexus: Native Mesh Generation with Diffusion