$C_4$-Equivariant Flow Matching on Anisotropic Power-Diagram Graphs for Microstructure Generation

Key points

• Acquiring realistic microstructure data through Electron Backscatter Diffraction (EBSD) is costly and time consuming, often relying on specialised equipment.1
• As microstructures strongly influence material properties, generating realistic samples is essential for modelling the behaviour of polycrystalline materials.1
• By representing microstructures as anisotropic power diagrams, our model learns a compact geometric parametrisation and can render generated samples at arbitrary pixel resolution.1
• A $C4$-equivariant architecture incorporates rotational symmetry directly into the model, ensuring that rotations of the input noise produce corresponding rotations of the generated microstructure.1

Sources (1)

• [1]$C_4$-Equivariant Flow Matching on Anisotropic Power-Diagram Graphs for Microstructure Generation
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 09:14 AM

  We introduce a generative model for synthesising realistic polycrystalline microstructures using flow matching and graph neural networks.

  Acquiring realistic microstructure data through Electron Backscatter Diffraction (EBSD) is costly and time consuming, often relying on specialised equipment.

Extractive summary: sentences quoted from the sources.