The sky continues to be the troublesome part to build a complete globe. Is there an easier way to just "fill-it" with something so the missing parts don't come out black?
> The sky continues to be the troublesome part to build a complete > globe. Is there an easier way to just "fill-it" with something so > the missing parts don't come out black?
> The sky continues to be the troublesome part to build a complete globe.
> Is there an easier way to just "fill-it" with something so the missing
> parts don't come out black?
Use the GIMP resynthesizer plug-in: http://registry.gimp.org/node/25219 If you have Photoshop, the context-aware fill may do the same thing but
I haven't tested that.
On Wednesday, May 23, 2012 3:19:54 PM UTC-7, markku...@iki.fi wrote:
> 24.5.2012 0:03, ecs1749 kirjoitti: > > The sky continues to be the troublesome part to build a complete globe. > > Is there an easier way to just "fill-it" with something so the missing > > parts don't come out black?
> Use the GIMP resynthesizer plug-in: http://registry.gimp.org/node/25219 > If you have Photoshop, the context-aware fill may do the same thing but > I haven't tested that.
On Wednesday, May 23, 2012 3:11:43 PM UTC-7, Bruno Postle wrote:
> On Wed 23-May-2012 at 14:03 -0700, ecs1749 wrote: > > The sky continues to be the troublesome part to build a complete > > globe. Is there an easier way to just "fill-it" with something so > > the missing parts don't come out black?
On Wednesday, May 23, 2012 5:23:41 PM UTC-7, ecs1749 wrote:
> What does it mean by using Mask to fill-it?
> On Wednesday, May 23, 2012 3:11:43 PM UTC-7, Bruno Postle wrote:
>> On Wed 23-May-2012 at 14:03 -0700, ecs1749 wrote: >> > The sky continues to be the troublesome part to build a complete >> > globe. Is there an easier way to just "fill-it" with something so >> > the missing parts don't come out black?
On Wednesday, 23 May 2012 at 23:11:43 +0100, Bruno Postle wrote:
> On Wed 23-May-2012 at 14:03 -0700, ecs1749 wrote:
>> The sky continues to be the troublesome part to build a complete
>> globe. Is there an easier way to just "fill-it" with something so
>> the missing parts don't come out black?
Interesting. That's very helpful. I didn't know you could add an
image more than once. I've tried this on some problem panoramas that
I have been working on, and it works well. Is there a reason why you
put this on Flickr rather than on the panotools Wiki?
I had difficultly with a few points, in particular how to get the
images where you want. Clearly you can't attach any control points,
because the areas don't match. Here's what I did:
1. Load all the images once and create control points normally.
2. Mask out the areas I didn't want.
3. Locate an image with a similar area and load it. Mask out the
parts of that image that I didn't want.
4. In the Images tab, select the image and set yaw to an appropriate
value, mainly trial and error.
That worked. My results, both before and after, are at
http://www.lemis.com/grog/photos/Photos.php?dirdate=20120510 There are
two views here. RCC-1.orig is stitched with no correction, RCC-1-1
was an attempt with GIMP, and RCC-1-2 is filled according to your
method. The same goes for RCC-2, RCC-2-1 and RCC-2-2. You'll note
that in this case I was able to get rid of the tripod shadow too.
A couple of problems: the Move/Drag functionality in the fast panorama
preview doesn't move the additional images, so it seems that you need
to move things into place first. Also reloading the .pto files puts
the additional images in different places.
So, my question: is this the correct way of doing it? Or is there an
easier way?
Greg
--
Sent from my desktop computer
Finger g...@FreeBSD.org for PGP public key.
See complete headers for address and phone numbers.
This message is digitally signed. If your Microsoft MUA reports
problems, please read http://tinyurl.com/broken-mua
There is a Windows plugin link lower in the page, search for "There is a
Windows version". I tested, it works with Gimp 2.8. Well, at least it does
not hang. But I guess finding how to use it may take some time as there
does not seem to be any documentation...
> Has anybody compiled this plug-in for Windows? Thanks.
> On Wednesday, May 23, 2012 3:19:54 PM UTC-7, markku...@iki.fi wrote:
>> 24.5.2012 0:03, ecs1749 kirjoitti:
>> > The sky continues to be the troublesome part to build a complete
>> globe.
>> > Is there an easier way to just "fill-it" with something so the missing
>> > parts don't come out black?
>> --
> You received this message because you are subscribed to the Google Groups
> "Hugin and other free panoramic software" group.
> A list of frequently asked questions is available at:
> http://wiki.panotools.org/Hugin_FAQ > To post to this group, send email to hugin-ptx@googlegroups.com
> To unsubscribe from this group, send email to
> hugin-ptx+unsubscribe@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/hugin-ptx
-- Frederic Da Vitoria
(davitof)
Membre de l'April - « promouvoir et défendre le logiciel libre » -
http://www.april.org
> Interesting. That's very helpful. I didn't know you could add an
> image more than once. I've tried this on some problem panoramas that
> I have been working on, and it works well.
You can fill in any old thing. I've sometimes even just filled in
gradients, or put in bits of images with false lens parameters to
stretch them. This kind of cheat is particularly good for zenith and
nadir. The only problem I sometimes get is that if the patch is too
different from the place where it's put, enblend fails to find a seam.
The images don't all move together when you drag them in 'standard'
mode, but I found a trick: if you move in 'standard, single images' or
whatever it's called in english, and then check all images, you can
move them together even if they aren't linked with CPs. And just
moving the whole panorama by entering yaw, pitch or roll values at the
top of the preview window also moves all images.
> I had difficultly with a few points, in particular how to get the
> images where you want.
Hi Greg, you can go to the Fast Preview Panorama window and choose the
Move/Drag tab. There you can drag the new image to the place you want. With
other mouse buttons you can rotate. I think that is the easier way to do
that.
> Use the GIMP resynthesizer plug-in: http://registry.gimp.org/node/25219 > If you have Photoshop, the context-aware fill may do the same thing but
> I haven't tested that.
Another option is to use the clone tool on gimp or PS.
Another one is to make a selection, feather it (on GIMP -
Selection/Feather) copy and paste over the black place.
I did some videos here using gimp to solve some minor problems on panoramic
images:
They don't have the quality I wish, but I guess they can illustrate some
possibilities. Sorry also for my English and maybe the link won't be the
best.
> Interesting. That's very helpful. I didn't know you could add an
> image more than once. I've tried this on some problem panoramas that
> I have been working on, and it works well. Is there a reason why you
> put this on Flickr rather than on the panotools Wiki?
Just lack of time, since flickr is quicker. I should move it to the
wiki but need to find the time. There are one or two other
mini-tutorials that ought to be transferred: one on editing panoramas
with panomatic and another on stitching a street elevation with the
mosaic functionality.
> Just lack of time, since flickr is quicker. I should move it to the
> wiki but need to find the time. There are one or two other
> mini-tutorials that ought to be transferred: one on editing panoramas
> with panomatic and another on stitching a street elevation with the
> mosaic functionality.
For the time being you could add a link in the tutorials list on the wiki...
On Thursday, 24 May 2012 at 16:24:48 +0100, Bruno Postle wrote:
> On 24 May 2012 07:33, Greg 'groggy' Lehey <groog...@gmail.com> wrote:
>> On Wednesday, 23 May 2012 at 23:11:43 +0100, Bruno Postle wrote:
>> Interesting. That's very helpful. I didn't know you could add an
>> image more than once. I've tried this on some problem panoramas that
>> I have been working on, and it works well. Is there a reason why you
>> put this on Flickr rather than on the panotools Wiki?
> Just lack of time, since flickr is quicker. I should move it to the
> wiki but need to find the time.
Greg
--
Sent from my desktop computer
Finger g...@FreeBSD.org for PGP public key.
See complete headers for address and phone numbers.
This message is digitally signed. If your Microsoft MUA reports
problems, please read http://tinyurl.com/broken-mua
first, thank you for the tutorial. I have already shown Bruno's before to
some people, now there is one more option.
I found a little strange your hands used to cover the sun. I might suggest
that you do many different expositions and combine them to make the sun and
the rest of the scene look ok. I am still getting better on doing this and
surely Magic Lantern is making this job easier, but I think it is possible
to do it manually also. Here is an example:
where I have used enfuse to combine 5 JPGs with 2EV of exposition
difference between each one. In this case it would be very difficult to
remove the sun in the middle of the trees.
I did the same using 3 exposures made with the AEB from the camera - no
Magic Lantern. And it is also possible to do with an only raw file. In this
case I suggest you make darker pictures to make them brighter manipulating
the raw file.
Hugin can do the exposure fusion, but in general I do this before stitching
the panorama. I test the enfusion and then make a shell script to repeat
the process with all images, then put the resulting images to be stitched
into hugin.
> On Thursday, 24 May 2012 at 16:24:48 +0100, Bruno Postle wrote:
> > On 24 May 2012 07:33, Greg 'groggy' Lehey <groog...@gmail.com> wrote:
> >> On Wednesday, 23 May 2012 at 23:11:43 +0100, Bruno Postle wrote:
> >> Interesting. That's very helpful. I didn't know you could add an
> >> image more than once. I've tried this on some problem panoramas that
> >> I have been working on, and it works well. Is there a reason why you
> >> put this on Flickr rather than on the panotools Wiki?
> > Just lack of time, since flickr is quicker. I should move it to the
> > wiki but need to find the time.
> Greg
> --
> Sent from my desktop computer
> Finger g...@FreeBSD.org for PGP public key.
> See complete headers for address and phone numbers.
> This message is digitally signed. If your Microsoft MUA reports
> problems, please read http://tinyurl.com/broken-mua
On Friday, 25 May 2012 at 9:39:37 -0300, Carlos Eduardo G. Carvalho (Cartola) wrote:
> I found a little strange your hands used to cover the sun. I might
> suggest that you do many different expositions and combine them to
> make the sun and the rest of the scene look ok. I am still getting
> better on doing this and surely Magic Lantern is making this job
> easier, but I think it is possible to do it manually also. Here is
> an example:
Yes, but this one (although a very nice photo) has internal
reflections from the sun. The reason I cover the sun with my hands is
to get rid of these reflections. I take two photos of the same view,
one with my hand in front of the sun, one without. The one with my
hand in front has no obvious flare or reflections, and I can mask out
only the sun from the one without the hand. Take a look at
http://www.lemis.com/grog/photography/sun-masking.php for details.
In the example I showed yesterday, I made a mistake: I forgot to take
one of the images without my hand, and this was 40 km from home.
That's why I had to retouch the panorama. But I think the results are
good enough that I might use them more often.
> where I have used enfuse to combine 5 JPGs with 2EV of exposition
> difference between each one. In this case it would be very difficult
> to remove the sun in the middle of the trees.
Agreed. I've had that problem too, though I don't see that the merged
images make a difference. In fact, the images in the page above were
also merged in that manner.
> I did the same using 3 exposures made with the AEB from the camera -
> no Magic Lantern. And it is also possible to do with an only raw
> file. In this case I suggest you make darker pictures to make them
> brighter manipulating the raw file.
I've been experimenting with HDR-like techniques on the one hand and
with extracting shadow detail from the raw image on the other hand.
I'm not really happy with either. Exposure bracketing gives good
shadow detail, but it's far too windy here to use it as a general
technique. I use DxO Optics "Pro" to extract shadow detail from the
raw images, but it can only get out what's there in the first place,
and the shadows tend to be washed out. I'm currently experimenting
with uneven exposure of each image, which seems to be giving me the
best results.
> Hugin can do the exposure fusion, but in general I do this before
> stitching the panorama. I test the enfusion and then make a shell
> script to repeat the process with all images, then put the resulting
> images to be stitched into hugin.
Yes, I've done that too in the past. Clearly it's not as easy with
the situation we're talking about.
Greg
--
Sent from my desktop computer
Finger g...@FreeBSD.org for PGP public key.
See complete headers for address and phone numbers.
This message is digitally signed. If your Microsoft MUA reports
problems, please read http://tinyurl.com/broken-mua
On Sat 26-May-2012 at 12:32 +1000, Greg 'groggy' Lehey wrote:
>Yes, but this one (although a very nice photo) has internal
>reflections from the sun. The reason I cover the sun with my hands is
>to get rid of these reflections. I take two photos of the same view,
>one with my hand in front of the sun, one without. The one with my
>hand in front has no obvious flare or reflections, and I can mask out
>only the sun from the one without the hand.
I do something similar, though I shoot one with my left hand covering the sun then another with my right hand. The result after masking out my arms, is a small hole where the sun should be and no lens flares. I do the same in the opposite direction to hide my shadow, the result is just the shadow from the camera as if it was floating in space.
This description would be better with some illustration..
On Sat 26-May-2012 at 21:50 +0100, Bruno Postle wrote:
>I do something similar, though I shoot one with my left hand covering >the sun then another with my right hand. The result after masking >out my arms, is a small hole where the sun should be and no lens >flares. I do the same in the opposite direction to hide my shadow, >the result is just the shadow from the camera as if it was floating >in space.
>This description would be better with some illustration..
Here is an example of an unmasked panorama, you can see the hands covering the sun and the shadow of the camera:
