Luminous Landscape Forum
Raw & Post Processing, Printing => Adobe Lightroom Q&A => Topic started by: Chris Kern on September 02, 2015, 11:58:49 pm
-
I've mostly avoided panoramas except for a few simple experiments because I assumed that to get good results I would need (1) a tripod with a specialized pano head and (2) dedicated software. But my wife and I were in Vancouver last week and I figured I had nothing to lose by trying to see what I could do shooting handheld and using Lightroom to combine the images.
I was pleasantly surprised by how well they turned out.
It appears LR's stitching algorithm can accommodate a lot of overlap. The attached cityscape was a composite of 21 (sic) images, although I've cropped a bit from both sides in this JPEG. I wasn't counting as I panned the camera: the "Lookout Tower" in the Harbour Centre building where I was shooting was filled with other tourists and I was too busy dodging them and making certain I kept the horizon roughly in the same part of the frame to keep track of how many times I snapped the shutter. When I got back to our hotel room and realized how many shots I had made, I thought I might have to cull some of the redundant ones before making the stitch, but LR managed the excessive overlap with no difficulty. The skyline in the second attachment, which I made the following day on the other side of the harbor (from Lonsdale Quay in North Vancouver, for those familiar with the city), is a composite of eight images. I was less distracted when I shot this series and each of the individual images only overlaps by about a third with its successor.
Both panos were shot handheld with a Fuji X-T1 and a kit 18-55mm lens. All the work in post was done on the LR-generated composites except for a little content-aware fill in Photoshop, so getting the final images to look the way I wanted involved no more effort than it would if the source had been a single raw image.
-
Nice shots....and yes, Pano Merge works really well.
-
I'm impressed, my own efforts were much less successful so perhaps the secret is to take plenty of shots rather than just the three I used. I shall try again.
-
I'm an experienced "panoramist", having shot and sold many full 360*180 panos, using the required pano head and sophisticated software in the past ten years.
That said, I just LOVE the impromptu panos I can do with Lightroom. A few quick points:
1/ Overlap is GOOD - and even needed so that the software understands and corrects for distorsion. From one image to the next, I typically aim for a 30% overlap (rule of thumb: the significant landmark on the right edge of picture n should move to the left edge of picture n+1, assuming you're panning left to right). You're not making thing difficult for LR if you have overlap, quite the opposite.
2/ Shoot in portrait mode for more coverage.
3/ For landscapes with no very close object, you definitely DON'T need a pano head, as any parallax error will be negligible. Indoors is definitely another story, and a full 360*180 will almost always require a pano head.
Best,
Stephane
-
Love the Pano tool in LR CC/6 too. Much better than the X-T1 in camera panos.
The resulting DNG files can be quite large, 21 16Mpx will get big. I've found that converting them to lossy DNG will reduce their size substantially. Unless you are printing wall murals with the files, you probably won't see the artifacts, I don't.
-
Both are very good CC's ability to use content aware to close up white space after the warp during the conversion is a neat addition LR can not do currently. I have also found the LR seems to work best in raw files not imported tifs.
Paul
-
I too love the convenience of doing panos in Lightroom, when it works. I've probably done 25 or 30 panos since upgrading to LR6, mostly on tripod, and Lightroom has worked on all but 4. On those 4, Panorama Factory successfully stitched the exported TIFFs without issue.
Dave
-
I have done a bunch of panos in LR and the smae images in CC. The end product from each is not the same. The LR merges seem to be a bit elongated and distorted, while the CC process gave me so such distortions. In addition, the Auto edge fill in THe CC process is very nice. It does add another step if using CC but I like the results better and will stick with this method.
MDIJB
-
I have also found the LR seems to work best in raw files not imported tifs.
I haven't performed any tests of creating panoramas from raw vs. cooked files, but Lightroom's ability to work with the former is a big win, as far as I'm concerned. Among other things, LR emits what I presume is a linear DNG: a demosaiced file, which I believe Eric Chan refers to as a "scene-referred" as opposed to "output-referred" format, that still permits non-destructive color adjustments. It's like working with a single raw capture file.
-
I haven't performed any tests of creating panoramas from raw vs. cooked files, but Lightroom's ability to work with the former is a big win, as far as I'm concerned. Among other things, LR emits what I presume is a linear DNG: a demosaiced file, which I believe Eric Chan refers to as a "scene-referred" as opposed to "output-referred" format, that still permits non-destructive color adjustments. It's like working with a single raw capture file.
I agree a huge benefit.
Paul
-
I agree a huge benefit.
Hi Paul,
I wouldn't call it a benefit, but rather a huge convenience.
Fact is, that a properly executed stitch with e.g. PTGUI, will potentially deliver a higher image quality. That's a benefit in my book, but it's less convenient to get there from a workflow perspective. The hugely larger level of control of a dedicated stitching program, and the better resampling quality, can deliver much higher image quality and make some types of stitches possible that will fail in LR.
What certainly is a benefit of the LR capability to stitch Raw files is that, with a lower threshold and added convenience, people will start using stitching more often than they did because they thought it was difficult and is impossible without purchasing expensive accessories.
Cheers,
Bart
-
I'm curious. Has anyone tried Lr's panorama as a quick and dirty focus stacking technique? I tried it and didn't get acceptable results, even though stacking with the same photos worked in Ps. I know that in Ps you have to select between a stack and a pano when blending, so I guess that's the difference.
But maybe if there was less overlap? or some other way of composing before using pano?
It would be great if in addition to the HDR and pano we had stacking.
-
I'm curious. Has anyone tried Lr's panorama as a quick and dirty focus stacking technique?
Nope...won't work but focus stacking is on Adobe's agenda for ACR & LR.
-
... a properly executed stitch with e.g. PTGUI, will potentially deliver a higher image quality...
+1
-
Nope...won't work but focus stacking is on Adobe's agenda for ACR & LR.
+25 ;D
-
Glad to hear that Adobe are working on focus stacking in LR.
I like the 2 photos but there's a (to me) whopping discontinuity in the 1st. The yellow rail along the bottom over to the right. I'm not sure but further right also looks suspicious.
I use Auto Pano and one of the things I like about it is that it does allow an interaction after putting together the image, and shows you a fit value, which if poor you can then troubleshoot (and potentially fix) with control points between the offending files. I'm pretty sure PTGUI offers a similar interaction. Having said that I've use LR pano, especially to generate some quick small jpegs and it does reasonable well.
-
I like the 2 photos but there's a (to me) whopping discontinuity in the 1st. The yellow rail along the bottom over to the right. I'm not sure but further right also looks suspicious.
That railing, and the railroad tracks below it, inauthentically appear to be bowed toward the camera—the result of my using LR's cylindrical projection to stitch the composite image. (The spherical projection exhibited similar distortion and the perspective projection was not available, probably because the arc was too wide: it covers the better part of 180 degrees, as best I can determine from a Google Maps satellite view.) I don't know whether another application could have maintained the straight appearance of those features near the center without significantly distorting those on the left and right sides; I don't have any experience with dedicated panorama-generation apps. I suspect, based on some experiments a while ago, that Photoshop would have produced a similar result. I presume there are some inherent limits to how much geometrical magic you can perform when you map a cylinder onto a plane.
-
Chris,
Were the singles taken vertically or horizontally?
Having used Ptgui and Autopano for years now, I personally don't agree that they can make a better image. Where both Autopano and Ptgui break down IMO is in the exposure blending. Often times both tools can't get a sky even especially a sky where the light is fading across the frame.
There is more user interaction available with both Ptgui and Autopano, but I have never really been able to do much with it. Autopano has some good video's but even then when you start really getting into the techno parts of the shot it can be very difficult at least to me, more time than it's worth, and then after all is said and done you don't get a good blend. And with Ptgui, that's using the enblend plugin. On the flip side, with CC panoramas, at least with Fuji X-trans tifs, the final output to me was always softened unlike Ptgui and Autopano. I had switched over to those tools for Fuji files due to the softness problems with CC. Now that it can all be done as a dng inside of LR, that issue is gone.
To me it's a balance of time needed and what's the net output going to be worth, as more and more folks are willing to accept a pano from the iPhone as top end.
What LR has to me is a better overall solution, sure it could use some fine tuning, and maybe that will come later. But the ability to have the image as a dng and then work it with the LR toolset is worth a lot.
The rail bending in the foreground is possibly also due to the lens, I would need to see the original images. But there has to be some curve there or maybe it's a 90 degree bend? The railroad track also pick it up to a degree. Some of this can be fixed by warping the final image in CC or using the Len correction tool on the final shot.
I don't think that a nodal solution would have fixed the curving of the rail. The only solution I can think of would be a shift lens setup.
LR and CC both to offer a better exposure blending in most cases than either Autopano or Ptgui, where the later are going to give you more options to fine tune the distortions.
I have been a stitcher since 2002 or so always to gain more overall resolution, not as much for the pano look. But the tools that are available now are excellent.
Paul
-
I like the 2 photos but there's a (to me) whopping discontinuity in the 1st. The yellow rail along the bottom over to the right. I'm not sure but further right also looks suspicious.
That railing, and the railroad tracks below it, inauthentically appear to be bowed toward the camera—the result of my using LR's cylindrical projection to stitch the composite image. (The spherical projection exhibited similar distortion and the perspective projection was not available, probably because the arc was too wide: it covers the better part of 180 degrees, as best I can determine from a Google Maps satellite view.) I don't know whether another application could have maintained the straight appearance of those features near the center without significantly distorting those on the left and right sides; I don't have any experience with dedicated panorama-generation apps. I suspect, based on some experiments a while ago, that Photoshop would have produced a similar result. I presume there are some inherent limits to how much geometrical magic you can perform when you map a cylinder onto a plane.
Sorry, I misunderstood your post the first time. You're quite correct: the roof edge with the railing in the center foreground is not cleanly stitched. I don't know whether that is the result of my sloppy handheld technique (I was maneuvering around a stream of other tourists as I made the individual shots) or some failing of Lightroom.
-
Were the singles taken vertically or horizontally?
Vertically.
To me it's a balance of time needed and what's the net output going to be worth, as more and more folks are willing to accept a pano from the iPhone as top end.
What LR has to me is a better overall solution, sure it could use some fine tuning, and maybe that will come later. But the ability to have the image as a dng and then work it with the LR toolset is worth a lot.
It certainly is to me. And I evaluate the cost-benefit ratio the way you do—although I also appreciate the desire of the pano perfectionists to achieve an optimal result.
I should reiterate that I shot far more individual frames than were necessary—or probably desirable—when I made that first image. I was navigating along the perimeter of the observation tower around many other tourists, who probably thought the guy snapping picture after picture with his camera in portrait mode was a little loopy, and I was paying as much attention to avoiding collisions as I was to the viewfinder. (Maybe if I had acted a little loopier, they would have given me a wide berth and made my job easier. . . . ) When I got back to the hotel and realized there were 21 shots in that sequence, my first thought was that I needed to cull out the more egregiously redundant ones or Lightroom was going to experience massive indigestion trying to combine them. But I decided to try using them all, just to find out what would happen. And what happened is what you see in the attachment to my original post.
-
Sorry, I misunderstood your post the first time. You're quite correct: the roof edge with the railing in the center foreground is not cleanly stitched. I don't know whether that is the result of my sloppy handheld technique (I was maneuvering around a stream of other tourists as I made the individual shots) or some failing of Lightroom.
Yes, that's it. I'm also suspicious of the railing at the end on the right. It appears to end in thin air.
I wonder if you have too many overlaps as you mentioned. It may be worth looking at the files and pulling one or more out in those spots if that's the case. I don't know about PTGUI but Auto Pano is available as a 30 day trial and may be worth playing with if you have any plans to do something with the image.
-
I wonder if you have too many overlaps as you mentioned. It may be worth looking at the files and pulling one or more out in those spots if that's the case. I don't know about PTGUI but Auto Pano is available as a 30 day trial and may be worth playing with if you have any plans to do something with the image.
I'm not sure how much time I want to invest in that pano—I actually like the skyline pano better as an image—but both of your suggested approaches are worth trying when I get some time.
By the way, I just checked, and PTGui also has a free trial.
-
I forgot to mention, that to my eye, I like the effect as it really creates a flow to the image. The building are nicely rendered, straight lines to the curves of the tracks and railing.
Paul
-
I forgot to mention, that to my eye, I like the effect as it really creates a flow to the image. The building are nicely rendered, straight lines to the curves of the tracks and railing.
Interesting you should mention that. I've encountered a number of instances—to my eye, at least—where some geometric distortion actually improves the appeal of an image. For example, I've found that if I correct all the keystoning in a shot of a tall building, I lose some of the sense of its height. I often like the photo better if I allow the top to converge just a little.
-
For those of you following this thread who are interested, I've attached full-resolution crops of the two stitching errors Mike Broomfield pointed out.
-
The 2nd one is a bit more critical. The first one if I am seeing the right spot is an easy fix.
The 2nd one would mean grabbing the correct part from one of the images, copying and pasting into the image, then warping that selection. Many times, you can fix such issues, can't tell on this one. If not, it's rework from the start.
Paul
-
The 2nd one is a bit more critical. The first one if I am seeing the right spot is an easy fix.
The 2nd one would mean grabbing the correct part from one of the images, copying and pasting into the image, then warping that selection. Many times, you can fix such issues, can't tell on this one. If not, it's rework from the start.
But there isn't much one can rework with, on an automatic stitch ... There are no controls in LR to do a better job, are there? Sure, postprocessing is always an option, but creating more work to correct such errors is not what stitching is supposed to do.
Cheers,
Bart
-
Personally my workflow is now LR first as I just prefer the dng output and ability to work the entire file as a raw. So if after a few attempts LR can't get it, depending on the error and scope I decide to either manually fix or move to ptgui and autopano.
Hopefully LR will improve over time but I doubt it will ever allow control point.
What a lot of folks don't remember is the old interactive CS3 option. I still fall back to that at times albeit only in 8 bit. That version allowed you to make changes to where rhe images were blended and many times a short tweak would fix a problem. And you could view the image at around 50 percent.
So farcical having very good success with LR. LR is even able to work out a solution for my old 3 part stitches with the Zork adapter which in the past always vexed most software as the software did not like the lens camera combination. Canon and Mamiya 35mm.
Paul
-
Does anyone know what is the resampling algorithm used for panoramas in LR6?
-
Does anyone know what is the resampling algorithm used for panoramas in LR6?
That is a good question, which I have wondered also, both with LR and CC. One aspect of Ptgui I have always liked is the ability to pick the algorithm for the warping.
I have assumed that LR uses the same algorithm that is used with standard uprez during export.
Paul
-
Fact is, that a properly executed stitch with e.g. PTGUI, will potentially deliver a higher image quality. That's a benefit in my book, but it's less convenient to get there from a workflow perspective. The hugely larger level of control of a dedicated stitching program, and the better resampling quality, can deliver much higher image quality and make some types of stitches possible that will fail in LR.
I've shot what I thought were easy panoramas that LR could not do by PTGUI did without a sweat. Go figure.
-
That is a good question, which I have wondered also, both with LR and CC. One aspect of Ptgui I have always liked is the ability to pick the algorithm for the warping.
I have assumed that LR uses the same algorithm that is used with standard uprez during export.
Paul
It is getting harder to tell what resampling algorithm is being used.
LR6 uses a different one than Bicubic Sharper for downsizing images, as far as I can tell no one yet figured what it really is.
Photoshop has been using bilinear (!) for lens corrections for quite a while too. I don't know if LR uses bilinear for lens corrections. Changing the default interpolator in Photoshop's settings does nothing to force lens corrections to say use something like bicubic. That's why lens corrections always resulted in rather blurred detail in PS!
-
It is getting harder to tell what resampling algorithm is being used.
Hi Samuel,
That's correct. Panorama stitching is often very much like distortion correction, some areas can get enlarged, others can get reduced in size, in the same image.
LR6 uses a different one than Bicubic Sharper for downsizing images, as far as I can tell no one yet figured what it really is.
Photoshop has been using bilinear (!) for lens corrections for quite a while too. I don't know if LR uses bilinear for lens corrections. Changing the default interpolator in Photoshop's settings does nothing to force lens corrections to say use something like bicubic. That's why lens corrections always resulted in rather blurred detail in PS!
LR uses a decent downsampling method (better than Bicubic and certainly better than Bicubic Sharper), although I do not know what it is exactly. The upsampling has some resemblance to bilinear. However, there may also be adaptive (to the image content) features at play. But the difficulty with stitching is the small rotations and enlargements/reductions. That will lose detail pretty quickly if the wrong method is used. Since LR's stitching is Raw input based, it's not easy to simulate with synthetic images which would show the trade-offs more quickly.
Dedicated stitchers like PTGUI could produce multiple conversions with the same pixel dimensions but interpolated/scaled with different filters, and the user can then blend the best parts manually, e.g. Lanczos2 for high contrast edges with sharp edge detail (yet with minimal halo), and a higher support version of Lanczos for structural detail, where ringing artifacts will be harder to see (if at all) due to the variable detail.
Cheers,
Bart
-
Hi Samuel,
That's correct. Panorama stitching is often very much like distortion correction, some areas can get enlarged, others can get reduced in size, in the same image.
LR uses a decent downsampling method (better than Bicubic and certainly better than Bicubic Sharper), although I do not know what it is exactly. The upsampling has some resemblance to bilinear. However, there may also be adaptive (to the image content) features at play. But the difficulty with stitching is the small rotations and enlargements/reductions. That will lose detail pretty quickly if the wrong method is used. Since LR's stitching is Raw input based, it's not easy to simulate with synthetic images which would show the trade-offs more quickly.
Dedicated stitchers like PTGUI could produce multiple conversions with the same pixel dimensions but interpolated/scaled with different filters, and the user can then blend the best parts manually, e.g. Lanczos2 for high contrast edges with sharp edge detail (yet with minimal halo), and a higher support version of Lanczos for structural detail, where ringing artifacts will be harder to see (if at all) due to the variable detail.
Cheers,
Bart
Hi Bart,
That's a nice way of looking at panorama stitching :-)
Yes, I recall reading your posts on downsampling interpolation in LR. How interesting that it is neither Bicubic nor Bicubic sharper. If I am not mistaken, in Photoshop, when Bicubic (automatic) is chosen, upsampling should use Bicubic Smoother. Strange that it resambles bilinear in LR!
I wonder if photographs of highly detailed targets could be captured, and then put through the paces of LR's stitching and PTGui, to reveal the trade-offs.
For my landscape pano stitches I am usually using Lanczos 8 in PTGui, with excellent results.
Interpolation to me is very important, I'm sad that it is not more vigorously discussed. I never saw it mentioned once in two recent lu-la articles on focus stacking. The whole point of that is to increase in-focus detail, so using an interpolator that makes everything softer is counter to the purpose. Unfortunately for the majority of the LR audience, a simple turn-key solution is preferable over fussing about such issues.
P.S. I had the opportunity to study at close range some billboard prints from Apple's 'Shot on iPhone 6' campaign, and they look surprisingly good. I wonder what upsampling software they used for those.
-
(http://www.pbase.com/dflynch/image/161028520.jpg)
I'm intrigued to read so many have had good success with PTGui. Probably I need to put more work into learning it, because on all my recent files with respect to distortion the initial output from Autopano is so much more pleasing to me. The above image is from 37 files initially output from SNS-HDR, using a tripod. With this image LR HDR + pano combo produced notably worse output.
The below image is from more than 200 handheld source files. LR CC very quickly ran out of memory and failed to create a pano; Autopano chugged away for a couple of hours and did the job. As I recall in PTGui the initial rendering of the horizon was distorted (i.e. not straight). The uneven sky comes only from using a polarizing filter.
(http://www.pbase.com/dflynch/image/161080473/original.jpg)
I had my tripod but I found rattling off the images handheld took far less time (helpful when you're on top of a hill with no toilet and a crook tummy ;D). Image quality subsequently does suffer in some places, but then again the air was so full of dust that it probably doesn't matter much in the overall scheme of things. Still some portions came out pretty well I think e.g.
(http://damonlynch.net/images/20150818-0751-304-iso320-f8.0-85mm-640-Panini-crop.jpg)
-
I'm intrigued to read so many have had good success with PTGui. Probably I need to put more work into learning it, because on all my recent files with respect to distortion the initial output from Autopano is so much more pleasing to me.
Hi Damon,
Autopano Giga is fine on large numbers of files if it actually works out automatically, however I find it cumbersome if manual adjustments need to be made. Also its colormanagement is a bit of a mystery to me. But it only uses a relatively limited number of projection methods, that are also available in PTGUI (and then some). So the perspective/projection distortion should look identical with the same projection selected in both stitchers.
The above image is from 37 files initially output from SNS-HDR, using a tripod.
Yes, that's also how I approach HDR scenery panos, first let SNS-HDR do its magic on the individual tiles, then stitch the tiles with e.g. PTGUI. Apparently you had too little space to back off a bit more, so you inevitably ended up with an extreme FOV which, in rectilinear projection, produces a somewhat stretched appearance.
The below image is from more than 200 handheld source files. LR CC very quickly ran out of memory and failed to create a pano; Autopano chugged away for a couple of hours and did the job. As I recall in PTGui the initial rendering of the horizon was distorted (i.e. not straight). The uneven sky comes only from using a polarizing filter.
In my experience, a distorted horizon can be simply solved in PTGUI by assigning a few horizontal control points to some of the images that are actually on the horizon. That would allow PTGUI to figure out the horizon's (Pitch) position in the overall scheme of things, and to level it at the same time. Polarization filters usually create problems is blue sky gradients on wide FOVs.
I had my tripod but I found rattling off the images handheld took far less time (helpful when you're on top of a hill with no toilet and a crook tummy ;D).
I can imagine. Distant scenes are relatively tolerant with stitching of handheld shots. Things get more critical with close up foreground detail or occlusions, and views through gates and doors/windows and such. Situations like you describe call for fast action, founded on experience. Practicing and experience does help.
Cheers,
Bart