Skip to main content


Did you know that @movingpandas also supports local image coordinates?

Currently working on a demo using #bicycle tracks extracted from #video footage 🚴πŸŽ₯ published by @mszll

https://arxiv.org/abs/2211.01301

#openscience #MobilityAnalytics #MobilityDataScience #mobility #MovingPandas #TrafficFlow #planning #cycling #TrafficBehavior
in reply to Anita Graser πŸ‡ͺπŸ‡ΊπŸ‡ΊπŸ‡¦πŸ‡¬πŸ‡ͺ

Very cool! Are you interested in the homography matrix, to be able to make the projection / correct speed calcs? I could be able to get it, potentially. (or at least the code to calculate it, from an older student project)
in reply to Anita Graser πŸ‡ͺπŸ‡ΊπŸ‡ΊπŸ‡¦πŸ‡¬πŸ‡ͺ

ah sorry, just realized, without the camera available anymore, we wont be able to calibrate and correct for the significant radial distortion, so I believe this wont be possible retrospectively πŸ™
in reply to Anita Graser πŸ‡ͺπŸ‡ΊπŸ‡ΊπŸ‡¦πŸ‡¬πŸ‡ͺ

I'm no expert on this, but I think the cam distortions would need to be fixed beforehand (can be a mix of radial+tangential), which seems practically impossible without cam calibration
in reply to Michael Szell

did some vector georeferencing with the #QGIS Vector Bender plugin https://plugins.qgis.org/plugins/VectorBender/ for fun to see where that would lead us

With enough ground control points, this could work to some degree. It won't be perfect but should improve speed estimates

#qgis #gischat #mobility #datascience #MovementDataAnalytics #cycling #MobilityAnalytics

Nicolas Roelandt :osgeo: reshared this.

This website uses cookies to recognize revisiting and logged in users. You accept the usage of these cookies by continue browsing this website.

⇧