Presentation

DiffusionNet: Discretization Agnostic Learning on Surfaces
SessionRoundtable Session: Neural Geometry Processing
Authors
Nicholas Sharp
Souhaib Attaiki
Keenan Crane
Maks Ovsjanikov
Presenter
Nicholas Sharp
Event Type
Technical Paper
Keywords
Modeling
Interest Areas
Research & Education
Presentation Types
In Person
Registration Categories
Full Conference Supporter
Full Conference
Exhibitor Additional Full Conference
Exhibitor Full Conference
Recordings
This session WILL NOT be recorded.
TimeTuesday, 9 August 20224:08pm - 4:16pm PDT
LocationEast Building, Room 1-3
DescriptionDiffusionNet is a general-purpose method for deep learning on meshes, points clouds, etc., via simple spatial diffusion. The networks generalize by construction across different samplings, resolutions, and representations of shapes and scale to high-resolution inputs. Spatial support is automatically optimized as a continuous parameter.
Authors
Nicholas Sharp
Carnegie Mellon University
Souhaib Attaiki
École Polytechnique
Keenan Crane
Carnegie Mellon University
Maks Ovsjanikov
École Polytechnique
Presenter
Nicholas Sharp
NVIDIA