Welcome, Guest. Please Login or Register
Home Help Search Login Register
Pages: 1 2 
Send Topic Print
painting my pano head (Read 11601 times)
Ex Member

Re: painting my pano head
Reply #15 - 06/15/10 at 04:03:09
So far I've been working on cleaning up the single nadir frame using 2 others shots taken at 90 and 180 degrees apart.
Once I have a clean frame, I use that as the nadir and then do the final stitch.

A little different approach.

great! I've got to try using the flexify method too.

However my question is in relation to painting on a HDR frame.

Would you drop down the exposure to a considerable level and then do the painting, alternatly raise the exposure to see how its shapping up in the highlights?

However with this process it takes twice the effort. Is there any other way you might know that cuts time?

I'm looking for a simpler workflow that would help shave off some time taken on each pano.

Such a very amazing link!
Thanks you for the post.

Watch Jonah Hex Online Free
Back to top
IP Logged
Ex Member

Re: painting my pano head
Reply #16 - 06/15/10 at 16:32:20
I use the following approach when stitching/patching the nadir with HDR workflow.  I use PTGui, which supports alpha channels in TIFF images.  I shoot RAW and do a RAW conversion to 16 bit TIFF.  Let's say I shot 6 exposures per pano segment.  I shoot three nadir shots:

1) with my pano head at 0 degrees (pan) I shoot straight down (6 exposures).

2) with the pano head at 180 degrees pan, I shoot straight down (6 exposures).

3) I move the tripod away from the nadir location (offset it)a couple of feet and tilt the tripod so that I am shooting at the nadir location from the new, adjacent area (6 exposures).

Once I do the RAW conversion on all of my files, I open the nadir images and mask the tripod out in the two straight down sets (put the mask in the alpha channel).  Because I used a tripod, the exposure set had the tripod in the same location, so it is simply a matter of copying and pasting the mask into the alpha channel of each of the six images.  In the attached images, RED represents the mask that I would apply to the alpha channel of each image.

For the offset nadir image, I mask everything that is not on the ground plane, leaving the ground plane visible.  Again, copy and paste the mask into the remaining images in the exposure sequence for the offset nadir image.

This is also a good time to rotate any of the nadir images that were output as landscape when what you need is portrait for PTGui.

Once this is complete, I bring all of the images into PTGui.  In this hypothetical example, I will bring in my 6 around + zenith + 2 nadir sets (0 and 180).  I will leave out the offset nadir for now.  Let PTGui do its thing and find automatic control points, etc.  Tweak the alignment to taste.  Because PTGui recognizes that the exposures were shot contemporaneously, it links them as HDR sets.  If not, enable this manually.  Because you masked the tripod out of the straight-down nadir shots, it should not appear in the panorama - however, there will be a cutout in the nadir area of the pano where the nadir is missing this info.  You patch it with the offset nadir image.

Now the final, offset nadir.  Import the offset nadir image set into the project and make sure they are linked for HDR.  Rerun the optimizer and you will get a poor fit - not to worry.  Now, manually add some control points to the offset nadir image across several of your 6 around and/or straight-down nadir shots so PTGui knows where to put the offset nadir shot.  You may also be able to help PTGui by manually dragging the offset nadir shot in the panorama editor window to get it close to where it should end up.

Now, go to the Optimizer and enable the "Viewpoint" correction checkbox for the offset nadir images - Viewpoint is the column next to the Roll column.

Run the Optimizer and you should get a perfectly patched nadir.  If not, you may need to add some more control points manually, or eliminate ones that laid in previously that may have been inaccurate.

Now you have a patched nadir within PTGui before any HDR combination and blending takes place.  No cloning necessary in post.

Please forgive me if this was already mentioned elsewhere, I don;t mean to be redundant.

The only extra work involves taking one more image set at the site, and the masking of the TIFFs prior to blending.  REMEMBER to shoot all of your other images prior to moving the tripod for the offset nadir!!!!

I usually do the masking prep, etc. on full-sized TIFFs and set the full-sized images aside. I make a reduced sized set of images with the same filenames (say 1024 or 2400 px on the long edge) for PTGui to chew on in terms of alignment and stitching. Once the stitch is done and appears accurate, I save the stitch as a template.  I then bring the full-sized images into a new project and apply the template.  This way the full-sized TIFFs don't need to be chewed through to determine the stitch, saving time and memory.  THis is especially useful on large, hi-res panos on my laptop.


Back to top

nadir-0.jpg (Attachment deleted)
nadir-180.jpg (Attachment deleted)
nadir-shifted.jpg (Attachment deleted)
pano-nadir.jpg (Attachment deleted)
IP Logged
Pages: 1 2 
Send Topic Print