Grouped CPs often indicate that the lense parameters are widely incorrect. On July 9, 2020 8:29:15 AM GMT+02:00, Frederic Da Vitoria <davito...@gmail.com> wrote: >Hello, > >I don't have an explanation, but I notice that Shaunak De's control points >are grouped instead of being spread on the common parts of the images. More >than the number of CPs, my guess is that the position of those CPs and the >fact none of them is far from the others is the cause of the problem. > >I suggest manually adding CPs as widely spaced as possible. I guess there >are too many comfusing things in those images. For example, if Hugin marks >a "e" on one picture, finding the matching "e" on the other could be >tricky, because there are a lot of exactly identical letters. So I'd put a >few carefully chosen CPs. For example, I'd put at least 2 CPs in the title >(one at each end of the common part) and one at the bottom of the column >which is barely visible at the bottom-left of the second image. I'd add a >few CPs as widespread as possible, which means probably a few close to the >center. Then I'd remove all the CPs which were automatically created, reset >all optiizations made in previous runs, and run optimization again. > >But this is a way to get around the issue, it does not explain why it >happened in the first place. > >Le jeudi 9 juillet 2020 02:19:25 UTC+2, Tduell a écrit : >> >> Hello Shaunak De, >> >> On Wed, 2020-07-08 at 16:41 -0700, Shaunak De wrote: >> > On further examination of your project file, I notice you have a lot >> more >> > control points than me. I can't seem to figure out why that would be. I >> am >> > using Hugin's default settings! >> > >> >> At the moment I can't think why that would be the case, maybe someone else >> will >> chime in with a possible explanation. >> I had a try with your full sized images, and that too worked OK. >> I didn't follow the tutorial explicitly, I used the following steps... >> >> 1. load images and apply 10 deg field of view to each, hugin automatically >> set a >> new lens for each. >> 2. run cpfind, which found 24 control points. A quick check of those >> looked OK. >> 3. Optimize in steps, to everything without translation. >> 4. check the result and adjust crop boundaries. >> 5. stitch. >> >> Attached is the pto file for this project. >> Hope that helps. >> >> Cheers, >> -- >> Terry Duell <tdu...@iinet.net.au <javascript:>> >> > >-- >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+unsubscr...@googlegroups.com. >To view this discussion on the web visit >https://groups.google.com/d/msgid/hugin-ptx/e00e42ba-d541-4305-bc96-9f364e33f71do%40googlegroups.com.
-- Sent from my Android device with K-9 Mail. Please excuse my brevity. -- 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+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/hugin-ptx/34A1030E-9AC6-44B9-A40C-BDB833E210F4%40gmail.com.
