2016-08-07 23:47 GMT+02:00 Michael Havens <havens.busin...@gmail.com>:
> I found the control points. Of the three pictures one of them would not do > it automatically so I inserted a couple manually. As for verifying the ones > it found automatically I didn't inspect them other than glancing at them > and seeing they appeared correct. > > How do I figure out lens distortion? Darktable knows my lens including > it's distortion. Can I use that information somehow? > > On Sun, Aug 7, 2016 at 12:52 PM, Sean Greenslade <s...@seangreenslade.com> > wrote: > >> On August 7, 2016 12:02:16 PM EDT, Michael Havens < >> havens.busin...@gmail.com> wrote: >> >Here are the photos and stitced picture. >> >https://drive.google.com/open?id=0B2xvsVTZy4y1T0NsMkdOdGo2TWs >> > >> >On Sunday, August 7, 2016 at 11:32:43 AM UTC-4, Michael Havens wrote: >> >> >> >> Hello gentlemen (and women), >> >> >> >> I take pictures of the interior of houses. I use hugin in lew of >> >> purchasing a wide angle lens (I do not have a spare $1000 dollars >> >laying >> >> around). I can only offer an idea, I don't have the technical >> >aptitude for >> >> anything else (wish I did). Now it is good but you can see the seams >> >if you >> >> look for them. What maybe someone could do is make it so control >> >points >> >> from images proportionally scale to fit together. Or perhaps I am >> >doing >> >> something wrong. >> >> >> >> There are many possibilities. If you took the photos handheld, or used a >> tripod without a nodal panning adapter, there could be uncorrectable >> parallax in the photos. Since there are objects in the near field, this is >> an important point to consider. >> >> Have you also gone through the entire optimization process? After finding >> control points, verifying that they are mostly all correct, then doing all >> the various optimization steps, including lens distortion. What was the >> final RMS error of the last optimizer run? >> > I am not sure you realize the importance of parallax errors. By visually comparing 222 with 223, I notice the chair does not cross the same part of the floor in the 2 pictures. I don't think any panorama software will be able to correct this. Same issue between 223 and 224, the intersection between the wheel and the seat behind it. You really need to take pictures so that those errors are as small as possible. The best way, of course, is to use a suitable tripod. At least, even if you don't or can't buy a tripod, you have to take care to turn your camera around itself instead of turning it around you. The second point to consider is the exposure issue. 222 was taken with completely different parameters (ISO 280) from the others (ISO 1800 and 2200). (BTW, I don't understand the parameters: 1/30, F3.5, I don't understand why the camera needs to go to ISO 280, I don't understand ISO 2200 more). In my experience, Hugin does not do a very good job about correcting different exposures like this. I suggest you should work on 222 to correct the exposure before trying to assemble. If you shot in RAW, you should be able to achieve a good correction. -- Frederic Da Vitoria (davitof) Membre de l'April - « promouvoir et défendre le logiciel libre » - http://www.april.org -- 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/CANe_y9QFPn0Grregt1V8MFY4XmEwRBE_q1reQbgv10kDjndiwg%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
