topoGenesis

Python Package of Computational Topology for Generative Systems and Sciences

August 20, 2020 - 1 minute read
Research Spatial Generative Python Genesis Lab Development

topoGenesis is an open-source python package that provides topological structures and functions for Generative Systems and Sciences.

Vision

topoGenesis aims to utilize the vast functionalities of fields (mathematical objects) in generative systems and sciences. Therefore it seeks to:

  1. offer basic mathematical functionalities on field data models
  2. offer functionalities of computational topology on top of the field structures
  3. facilitate the conversion between mesh-based data models and field data models.
  4. facilitate field simulations, whether governed by differential equations, spectral models or based on computational models (ABM)
  5. construct a bridge between spatial data models and tensor data structures to facilitate the utilization of the latest artificial intelligence models

Structure

  • Mesh to Field: Rasterization
    • Point Cloud Regularization
    • Line Network Voxelation
    • Mesh Surface Voxelation
    • Signed Distance Field
  • Field to Mesh: Isosurface
    • Boolean Marching Cubes
    • Marching Cubes
    • Surface Nets
  • Local Computation
    • Stencil / Kernels
      • von Neumann neighbourhood
      • Moore neighbourhood
      • Cube neighbourhood
      • Custom neighbourhoods
    • Universal Functions & Mathematical Operators (Numerical)
  • Field Simulations (Vectorized)
    • Dynamic Systems (based on Differential Equations)
    • Agent-Based Modeling
    • Cellular Automata

GSS-Lab

This project is currently being developed by Shervin Azadi and Pirouz Nouria at GSS-Lab, Department of Architectural Engineering and Technology, at TU Delft.