Hello everyone,
My questions are a follow up of this one:
https://support.capturingreality.com/hc/en-us/community/posts/360015325331-Apply-same-camera-distortion-from-one-image-to-another
This idea is to duplicate the XMP files associated to one frame from previously calibrated still videos to the other videos frames, in order to apply the same undistortion.
However, even if the XMP pose prior is “locked” and the calibration prior is “exact”, the new alignment is still requiring a lot a computation (about 15 min for 1k images). How so, since it should just load the parameters from the XMPs ? Is there a way to prevent any unnecessary computation ?
Even worse, the calibration and lens distortion parameters are not preserved. The difference in the calibration parameters is subtle (least significant digits). However, the difference in the lens distortion parameters is important and the resulting undistorted images do not match the original undistorted frame.
Am I doing something wrong ? Re using previous calibration parameters exactly should be quite straightforward and way faster than actually computing the alignment.
Many thanks,
Hi George,
15 minutes for 1000 images ad still not fast enough?!? 
Does that include feature detection? If so, then it’s fast enough in my opinion.
What kind of resolution are we talking about?
Prior only means something like preliminary, which in my understanding means it will be used as a starting point.
There should be an option somewhere that the source is undistorted - don’t have a license atm so can’t check…
Hi Götz,
My goal is to reuse the exact same calibration and lens distortion parameters. Therefore no computation should take place and resolution should not have any impact whatsoever.
Indeed, considering the current computation time, my guess is that RC still estimates new parameters and does the feature detection and matching - and would be fast enough if that was the actual task.
From the documentation, the prior being “locked” means “camera position and calibration will not be changed”, and is “fixed in further calculations”, so my understanding is RC should consider these cameras as already calibrated. But it is obviously not the case here, why !?
I am not sure I follow your last point. The source cameras are already calibrated, with parameters loaded thanks to XMP files from previous calibration. But the input images are still distorted at this point. The goal of this whole process is to undistort them according to the input lens distortion parameters.
Hi George,
asking for the resolution was rather to get a feel for the calculation time. If it’s 640x48, then 15 minutes might be much, if 50 mpx, then it’s very fast!
The features will be detected only once for each image as long as it’s not altered and the cache not emptied. In the component report it will give you more detailed info about how long each step takes.
Anyway, this isn’t your main point.
I think you’re not the first one to mention that “priors locked” doesn’t work as expected. I might confuse it with something else, but there might be an option somewhere where you can define an image as “undistorted”, which would in my opinion translate in “don’t touch it at all”.