Thank you so much for trying to make this work. I've had a couple of crises 
pop up the last few days and haven't been able to respond.

I think the lesson I learned is there are limits on what this tool can work 
with. There are really no sharp point on the near image other than the roof 
and there are certainly no sharp points on both images. Further, the snow 
around the trees makes it impossible to figure out what is in focus on the 
far image and I suspect that's why the ghosting around the trees shows up. 

On Thursday, April 10, 2025 at 8:21:01 PM UTC-6 GnomeNomad wrote:

On 4/10/25 14:39, Matt Rosing wrote:

Thanks. I was looking at that again and the close image really stands on 
its own. I think the combined image might be too cluttered. 

What I've managed to produce so far is too cluttered. The massed detail of 
the trees overpowers the icicles.


That said, I'm still trying to figure out how to use this tool. The images 
seem to be projected onto a sphere, then aligned, and then projected back 
to a plane. That makes sense if the images are part of a panorama but these 
images are mostly on top of each other and I had to do a bunch of roll, 
yaw, and pitch of the resulting image and I have no idea why.

First time, I tried using Hugin's align image stack function. That failed 
with an error. I think the focus in the two images is so far apart it can't 
find any control points between the two.

Second time, I loaded the images, went to the GL preview window, and ran 
the Assistant. It gave me an image with the icicles very sharp and the 
trees all blurry (because that's how they are in the close image), even 
using the Exposure fused from stacks output.

Third time, I gave the far image a second lens, ran the various alignment 
steps, and came up with - an image much like what I got the second time.

The fourth time, I added a mask to the close image, drawing one to include 
the biggest of the icicles. That produced an image with that icicle sharp, 
and nothing else improved over the previous images. While you could go 
through and mask everything you want included in each image (icicles from 
close, trees from far)... that's a lot of work and I'm not sure it would 
produce anything like what you want.

Fifth time, I removed the mask and set Hugin to output remapped images. 
Then I fed them to enfuse using a command line based on what I found at 
this URL:

https://macrocam.blogspot.com/2013/09/using-hugin-for-focus-stacking.html

That produced an image combining the focused trees with the focused 
icicles. Well, the icicles are sort of focused. There's an odd, ghost-like 
sort of look to them, like enfuse sort of fused the in-focus icicles with 
their out of focus versions. But it's enough to reinforce the likelihood 
that an image combining the two would be cluttered and crowded.

So I think it *could* be done, I don't think it should be done. The close 
image looks to me like a much stronger image than the far image. In the far 
image, the icicles just look like mistakes, that you were trying to get a 
picture of the sun-lit trees, and the icicles are just in the way.

I think this is one case where doing an actual painting combining the two 
images could make it work, but even then it might look cluttered.

The focus stacking might work better if you have more focus layers than 
just two. Perhaps images with the two dark stumps in the snow are in focus? 
Well, maybe not. I see they are in focus in the far image, while the trees 
themselves aren't as sharp.

Maybe you could try it with a pinhole could come up with enough depth of 
field to get it all in focus?


-- 
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 visit 
https://groups.google.com/d/msgid/hugin-ptx/7d682fdf-be28-44b8-9e41-ba75168bb12en%40googlegroups.com.

Reply via email to