/soft_object_perception

Primary LanguagePythonMIT LicenseMIT

Neural_soft_object_perception

Basics

Use MLS-MPM solver to simulate water/elastic/snow/sand/mud. Use render_particles.py to render the particles. To install taichi, use pip: python3 -m pip install taichi To run the demo, use python3 demo/demo_name.py

To simulate and render 3D scene (soft object colliding with an invisible object) with Python

  1. Make sure you have a modern NVIDIA GPU (e.g. GTX 1080 Ti)
  2. run python3 demo/object_collider_render.py, a binary file called sim_xxxx-xx-xx_xx-xx-xx will be created under the current folder (in this case, demo).
  3. To render the particles, run python3 your/directory/path/engine/render_particles.py your/directory/path/sim_xxxx-xx-xx_xx-xx-xx [begin frame] [end frame] [frame step] [render_output_folder]
  4. Images are now rendered and located in the [render_output_folder]