Sources for "Tracing real rays",
Revision 2020 (wild-category) entry by Nupo
This zip includes:
- Raw data captured from the GNSS-receivers (UBX-format) and the distance
measurement module. No data from the base as it's not needed here.
Receivers were configured to output only relative coordinates in
"RELPOSNED"-messages so there's no absolute position available.
- "movie script" that the Processing-sketch uses. Generated using the
"GNSS-stylus"-program (can be found from github)
- XYZ point clouds generated from the raw data (and tags)
- Obj-files generated from the point clouds (using MeshLab)
- Processing sketch (version 3.5.4 used) used to generate the CGI-parts of
the video
Running the processing sketch "ToiletSeat_Processing/ToiletSeat_Processing.pde"
should start rotating a toilet seat (and accessories) on your screen. There
are some swithes gathered to the top of the sketch that can be used to modify
the rendered scene, most influential ones being "cameraType" and
"scriptReplaySpeed".
File "Logs/Toilet_tags_original_unedited.tags" contains the original tags
added while "raytracing". It has more tags than the edited one. Data in the
end of the files is, however, corrupted. Battery of the laptop used for
scanning died and after restarting it using a car battery and inverter data
was no longer plausible. Probably caused by the inverter and laptop's power
supply being right next to the base's GNSS-receiver, causing too much noise.
Paper reel and brush were "raytraced" separately later (and in different
coordinates), no scanning data available for them.
