← Papers

PointNet CFD

Point-cloud deep learning for flow fields on irregular geometries

Paper title A Point-Cloud Deep Learning Framework for Prediction of Fluid Flow Fields on Irregular Geometries

Kashefi et al. (Stanford University) · Aero & CFD

Paper arXiv 2010.09469PDF

Takes the vertices of the CFD mesh as a point cloud and passes them through a PointNet segmentation network: each point gets its own features, a max-pooled global feature describes the whole shape, and the two together give velocity and pressure at every vertex. The data stay on the original unstructured mesh, fine near the body and coarse far away, with no interpolation to a pixel grid.

Published
Physics of Fluids

Share on LinkedIn, X or Bluesky, or or .

Know related work that belongs here?

Something wrong on this page?

More papers

All papers
  1. PhysicsNeMo-Mesh — GPU-Native Meshes and Simulation Cases for Physics AI
  2. Multi-fidelity scaling — Compute-optimal data scaling via multi-fidelity training
  3. CAE-UQ — Predictive Uncertainty for Neural CAE Surrogates