head-inthe-cloud/palms-plane-based-accessible-indoor-localization-using-mobile-smartphones ? reverse-engineered prompt

Build me a working Python project for indoor localization that matches this PALMS and PALMS+ repo.

I want to be able to run one version that uses LiDAR or ARKit style plane detections with IMU tracking, and another version that works from regular images by estimating depth first. It should take a floor plan and some example data, then estimate where the user is inside the building and show the result clearly with visualizations like a heatmap, point cloud, and a tracked path.

Please make the example scripts easy to run from the command line with config files, and include the sample data flow so I can try the demo without having to understand all the research details. If there are tests or basic checks for the main pipelines, keep those working too. Use current docs online if you need to confirm anything about the depth model setup or dependencies.