/pcd-lidar2video

Something something Oxford hackathon

Primary LanguagePython

pcd-lidar2video

The following open-source projects were forked to fit our data.

Build & run

  1. Install all dependencies (simply activate python venv with source bin/activate or equivalent). If running Windows and WSL2, plotting requires a X11 server on host (e.g. VcXsrv);
  2. Convert PCD file format to bin using src/pcd2bin.py;
  3. Pipe bin file to src/pcd-tools.py to plot scenes. For the hackathon this is fixed to an 8-frame range;
  4. Export desired scenes to PNG;
  5. Convert PNGs to m4v video using ffmpeg at 2 frames per second:
ffmpeg -start_number 1 -y -r 1 -i "video-%1d.png" -vf scale=1920:-1 -vcodec libx264 -b:v 10000k -r 1 outputhd.m4v

Route details

  • Location: University of Oxford's Computer Science Department (starting at the entrance to the University Parks);
  • Avg. speed: 29.5mph;
  • Timeline: Tue Jan 21 2020 11:20:35 GMT+0.

Render of the LIDAR data route taken.