Self-calibration code works fine for 3 cameras but not 5 cameras

[Gwo Jong Huang]

I have an instrument equipped with 3 cameras.  Each camera has fixed focal length (12.5 mm) and 2448x2048 2/3" CCD.  I used the self-calibration code (V1.0 without octave) to calibrate these 3 camera and got reasonable results.  No any warning show during the code running.  Although the principal points are off and focal length is from 13 mm to 12.1 mm.  I added 2 extra cameras to this instrument.  These two extra cameras also have 12.5 mm fixed focal length lens but with 1024x964 1/3" CCD.  It shows a lot of warning messages when I run the code for 5 cameras.  In the end, the shows large error and unreasonable results.  Does any one has same problem or have any idea what going wrong?

Thank you

Huang      

[Straw,Andrew]

Dear Huang,

In my lab, we usually find the results usually very good (with any number of cameras up to 11) but occasionally a particular camera configuration just seems to fail. Moving one of the cameras but just a small amount (a couple centimeters or so for a setup on a scale of 1 meter) often helped. I wonder if this could be related to the problems described here:

Oliensis, J., & Hartley, R. (2007). Iterative extensions of the Sturm/Triggs algorithm: convergence and nonconvergence. IEEE Transactions on Pattern Analysis and Machine Intelligence, 29(12), 2217–33. doi:10.1109/TPAMI.2007.1132

Best,

Andrew

--
You received this message because you are subscribed to the Google Groups "MultiCamSelfCal" group.
To unsubscribe from this group and stop receiving emails from it, send an email to multicamselfc...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

_______________________________________________

Andrew D. Straw, Ph.D.
Research Institute of Molecular Pathology (IMP)
Vienna, Austria
http://strawlab.org/