Tutorial for DNN Stereo Depth Estimation with Isaac Sim

https://media.githubusercontent.com/media/NVIDIA-ISAAC-ROS/.github/main/resources/isaac_ros_docs/concepts/stereo_depth/ess/Rviz.png/

Overview

This tutorial walks you through a graph to estimate depth with stereo images from Isaac Sim.

Tutorial Walkthrough

  1. Complete steps 1-7 listed in the quickstart.

  2. Install and launch Isaac Sim following the steps in the Isaac ROS Isaac Sim Setup Guide.

  3. Press Play to start publishing data from the Isaac Sim.

    https://media.githubusercontent.com/media/NVIDIA-ISAAC-ROS/.github/main/resources/isaac_ros_docs/getting_started/isaac_sim_sample_scene.png/
  4. Open a second terminal and attach to the container:

    cd ${ISAAC_ROS_WS}/src/isaac_ros_common && \
    ./scripts/run_dev.sh
    
  5. In the second terminal, start the isaac_ros_ess node using the launch files:

    ros2 launch isaac_ros_ess isaac_ros_ess_isaac_sim.launch.py engine_file_path:=/workspaces/isaac_ros-dev/src/isaac_ros_dnn_stereo_depth/resources/ess.engine
    
  6. Optionally, you can run the visualizer script to visualize the disparity image.

    ros2 run isaac_ros_ess isaac_ros_ess_visualizer.py
    
    https://media.githubusercontent.com/media/NVIDIA-ISAAC-ROS/.github/main/resources/isaac_ros_docs/concepts/stereo_depth/ess/Visualizer_isaac_sim.png/