Presentation

<
Generative Modeling with Orbit-Space Particle Flow Matching
SessionDifferentiable Geometry Processing & Flow Maps
DescriptionOrbit-Space Geometric Probability Paths (OGPP) is a flow-matching framework for generative modeling of particle systems that exploits shared geometric structure and uses probability path geometry to encode geometric information. It enables high-quality, efficient generation of particle configurations and point-cloud surfaces with normals, achieving state-of-the-art performance across diverse generation tasks.
Authors
Sinan Wang
Georgia Institute of Technology
Jinjin He
Georgia Institute of Technology
Shenyifan Lu
Georgia Institute of Technology
Ruicheng Wang
Georgia Institute of Technology
Greg Turk
Georgia Institute of Technology
Bo Zhu
Georgia Institute of Technology
Event Type
Technical Paper
TimeWednesday, 22 July 20262:50pm - 3:00pm PDT
LocationRoom 403 B
Digital Library PDF
Session Time & Location
Sunday, 19 July 20266:00pm - 8:45pm PDTHall K
Wednesday, 22 July 20262:00pm - 3:30pm PDTRoom 403 B
Keywords
Geometry
Modeling
Registration Categories
Full Conference Supporter
Full Conference
Next PresentationNext PresentationDifferentiable Geometry Processing & Flow Maps - Interactive Discussion
Similar Presentations
Neural Particle Automata: Learning Self-Organizing Particle Dynamics
A Few-Step Generative Model on Cumulative Flow Maps
Garment Particles: A 2D–3D Symmetric Garment Representation for Generation and Editing