Corpus ID: 3613313

Geodesic Convolutional Shape Optimization

  title={Geodesic Convolutional Shape Optimization},
  author={P. Baqu{\'e} and Edoardo Remelli and F. Fleuret and P. Fua},
  • P. Baqué, Edoardo Remelli, +1 author P. Fua
  • Published 2018
  • Computer Science, Mathematics
  • ArXiv
  • Aerodynamic shape optimization has many industrial applications. [...] Key Method The key to making this approach practical is remeshing the original shape using a polycube map, which makes it possible to perform the computations on GPUs instead of CPUs. The neural net is then used to formulate an objective function that is differentiable with respect to the shape parameters, which can then be optimized using a gradient-based technique. This outperforms state- of-the-art methods by 5 to 20% for standard…Expand Abstract
    17 Citations
    MeshSDF: Differentiable Iso-Surface Extraction
    • 2
    • PDF
    Learning three-dimensional flow for interactive aerodynamic design
    • 29
    • PDF
    Geometric Deep Learning for Volumetric Computational Fluid Dynamics
    DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation
    • 284
    • PDF
    Learning to Optimize Multigrid PDE Solvers
    • 9
    • PDF
    Deep Sketch-Based Modeling of Man-Made Shapes
    • 10
    • PDF


    Accelerating Eulerian Fluid Simulation With Convolutional Networks
    • 228
    • PDF
    Convolutional Neural Networks for Steady Flow Approximation
    • 135
    • PDF
    Learning three-dimensional flow for interactive aerodynamic design
    • 29
    • PDF
    Template-Based Monocular 3D Shape Recovery Using Laplacian Meshes
    • 36
    • PDF
    Adam: A Method for Stochastic Optimization
    • 52,682
    • PDF
    All‐Hex Mesh Generation via Volumetric PolyCube Deformation
    • 126
    • PDF
    • 225
    • PDF
    Response Surface Methods for Efficient Aerodynamic Surrogate Models
    • 4
    State-of-the-art in aerodynamic shape optimisation methods
    • 55
    • PDF