Reverse panoramic - emulating a fancy motorized star-tracking camera rig

[Benjamin Hill]

It feels like this should be possible, but I don't know if the math actually works out! 

Anyone interested in collaborating?

1. Take a lot of photos of the night sky with a fish eye camera.  (Don't feel bad if they are a noisy mess because your old GoPro has horrible low-light sensitivity.)
   
2. Track the brightest 10 stars using some very naive code (that I wrote) which dumps out 100k worth of alignment points in PTO format
3. Run autooptimiser on it and go get a coffee.
4. Run nona on the pano (more coffee)
5. Now you have a stack of 100 photos, all perfectly (?) rotated and un-distorted that you can sum or do a median filter on, and 
6. Get a single beautiful picture of the night sky rivaling those people with a huge lens and a motorized camera mount.

I got public code that works, and I think it does better than any other method I've seen - and doesn't require perfect knowledge of your camera's fisheye distortion.  Anyone want to collaborate and help make it less naive?

 

[Gnome Nomad]

Impressive! I'm not that much into astronomy shooting, but a friend of mine is, will forward it on to him.

--
A list of frequently asked questions is available at: http://wiki.panotools.org/Hugin_FAQ
---
You received this message because you are subscribed to the Google Groups "hugin and other free panoramic software" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hugin-ptx+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hugin-ptx/4cc92950-c1a7-445f-8170-cf5a4464e9e2%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Matthew Petroff]

Another way to align the images is to use the Astrometry.net software that will solve for the pointing of each image on the celestial sphere [1]. There's also a paper that uses pointings from said software to combine a large number of astronomy images to extract details that aren't visible in the source images [2].

-Matthew

[1] http://astrometry.net/
[2] http://arxiv.org/pdf/1406.1528v1