4d_recorder_sim

Introduction

4D Recorder Simulator based on NVIDIA Isaac Sim.

USD Files

• demo/ParticleInflatableDemo.usd

Getting Started

1. Install NVIDIA Isaac Sim

   sudo apt update \
     && sudo apt install libfuse2

   • check IOMMU

     sudo dmesg | grep -e DMAR -e IOMMU

2. Set the following environment variables to your ~/.bashrc file

   echo "export ISAACSIM_PATH='${HOME}/.local/share/ov/pkg/isaac_sim-2023.1.0-hotfix.1'" >> ~/.bashrc

3. Install conda

4. Create conda environment

   conda create -n sim python=3.10 \
     && conda activate sim \
     && cp -r conda_setup/etc $CONDA_PREFIX \
     && conda activate sim \
     && conda install tqdm \
     && pip install opencv-python

5. Verification

   python -c "from omni.isaac.kit import SimulationApp" \
     && python -c "import torch; print(torch.__path__)"

   • output

     ['~/.local/share/ov/pkg/isaac_sim-2023.1.0-hotfix.1/extscache/omni.pip.torch-1_13_1-0.1.4+104.2.lx64/torch-1-13-1/torch']

Build and Test (with the Isaac Sim path setup above)

Particle Inflatable Demo

• Camera cylinder set

python scripts/cylinder_camera_set.py usd/demo/ParticleInflatableDemo.usd

• Camera rectangle set

python scripts/rectangle_camera_set.py usd/demo/ParticleInflatableDemo.usd

Make dataset

• Camera cylinder set

./scripts/make_dataset.sh  cylinder

• Camera rectangle set

./scripts/make_dataset.sh  rectangle

Convert png to mp4

python scripts/convert_png_to_mp4.py data/isaac_sim/Collisiongroups_C9H3_661

Upload dataset

python scripts/upload_dataset.py isaac_sim Collisiongroups_C9H3_661