Ultimate R1 Calibration for a 5D2 w/Sigma 8mm

Welcome to Nodal Ninja Forum

+ Reply to Thread
Results 1 to 3 of 3

Thread: Ultimate R1 Calibration for a 5D2 w/Sigma 8mm

Share/Bookmark
  1. Ultimate R1 Calibration for a 5D2 w/Sigma 8mm

    #1

    I have a Canon 5D Mark II w/ Sigma 8mm and I have a stitching errors in my pano. The head that I am using is a Ultimate R1 that I bought about 2 months ago. I have attached a link at the bottom of the this post. There you will find a 360 that clearly has stitching issues, look at the nadir, above the couch and on top of the LCD in the pano and there you will find errors. I have been reading around and from what I have gathered I need to calibrate the R1. Problem is i don't really now how to do that. I dont think PTGui Pro is the problem either. I am new to panoramic photography (Newbie or Noob) but I know this much, I was unable to find any tutorials or reviews on the setup that I have.


    One more thing I bought and sent my Tokina 10-17 Lens off to be shaved by www.360pano.de/en/tokina-sigma-nikon.html (Thanks Tobias). I know that this lens will give the 5D2 justice! Will this remove my problem of having to calibrate my gear and will the R1 w/ring work on the Tokina 10-17?

    www.host.360twist.net/livingroom

    P.S. Don't be alarmed I am still working on my FPP template...

    Frankie
    www.360twist.com
  2. Re: Ultimate R1 Calibration for a 5D2 w/Sigma 8mm

    #2

    Frankie, The stitching errors may be due to the setup of the R1 head or the PTGui processing. I would be inclined to think that it's the latter that's responsible. PTGui uses control points assigned to matching features in overlapping images to align the images and also to envaluate lens parameters that are used to correct lens distortions. In order that the lens distortions can be adequately corrected, you need control points that are well spread out along the overlap areas. When PTGui generates the control points automatically, there is a tendency for them to be clumped together in the middle of the overlaps rather than nicely spread out - particularly in the case of fisheye lenses. What you may need to do is manually assign additional control points in the floor and ceiling areas, so that the optimizer has a sporting chance of correcting the distortion in those areas. So switch into Advanced mode via the button on the right of the Prioject Assistant tab, and select the Control Points tab to check the control point assignments. After assigning further points, select the Optimizer tab. In the Minimize Lens Distortion pull down menu, select Heavy + Lens Shift. The click Run Optimizer.

    Having done that, the stitching errors should be very much reduced. There may still be some inacurately placed control points, so it is worth clicking on the Control Points menu and selecting Delete Worst Points. Then run the optimizer again. Generally, you should aim to get the average control point distance down to <1 pixel ideally. The control points table can be displayed to reveal any remaining badly placed points. The worst ones (large distances) can be investigated and corrected by double clicking them, which takes you to the points on the Control Points tab. When the control point distances are finally very low, the images will align very well at the places where there are control points, and hopefully the stitching errors will be eliminated. A room with large expanses of featureless blank walls can present problems as there will be few, if any, features or control points. In those circumstances, people often resort to sticking a few postit notes in areas where images overlap to help in aligning the images. The notes can be cloned out of the final pinorama in Photoshop.

    You can check the head setup by including something like a lamp stand really close to the camera in a position where the images overlap. Then compare the overlapping images to see if the near object appears to shift relative to the background beween the two shots. (Obviously, don't have a featureless white wall in the background). Eliminate any parallax shift by adjusing the camera position on the head. See http://www.johnhpanos.com/epcalib.htm.

    John
  3. Re: Ultimate R1 Calibration for a 5D2 w/Sigma 8mm

    #3
    Users Country Flag
    Join Date: Apr 2008
    Location: Hong Kong
    Posts: 2,364

    Quote Originally Posted by 360twist View Post
    One more thing I bought and sent my Tokina 10-17 Lens off to be shaved by www.360pano.de/en/tokina-sigma-nikon.html (Thanks Tobias). I know that this lens will give the 5D2 justice! Will this remove my problem of having to calibrate my gear and will the R1 w/ring work on the Tokina 10-17?
    A lens ring consists of 3 components: outer metal ring, lens ring plate (or slider) and plastic inner ring. The inner ring is custom fitted to the contour of the lens. Unless 2 lenses share the same shape at the clamping position, they need different inner ring. We now have 20 different inner rings to fit more than 20 lenses! To serve these lenses, we have 4 outer rings of different sizes and 5 lens ring plates of different lengths.
    To cut the long story short, the outer ring is the same for Sigma 8mm Canon mount and the Tokina 10-17mm. Other parts need to be replaced. At the moment, we are not ready to sell the individual components.


    Nick




    Fanotec
    We listen. We try harder.
+ Reply to Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts