Tag Archives: processing

Lightroom’s New Panorama Feature

Lightroom CC comes with a number of new features. High on the list is the new panorama feature. Previously, if you wanted to stitch a panorama, you would select the shots in Lightroom, make sure you had them synced up for any exposure edits and white balance corrections and then send them to Photoshop. Photoshop had a Photomerge function that you could use or you could open them as layers in a document and carry the merge out yourself with a bit more control. This was the approach I used and I wrote about the technique here a while back.

The new Photomerge in Lightroom CC is designed to do without Photoshop. It takes your original files and then makes a new DNG file which is the stitched panorama but, as a RAW format, it still allows you to edit the image using the normal editing tools. I had to give this a go so, how well does it work?

The answer is a mixed one. In many cases, it works just fine. It stitches together the shot nicely and you can go on your way. While the initial preview process takes up a little time, the processing of the final stitch is done in the background so you can get on with something else – something I find very handy. However, it does have some shortcomings.

First, it doesn’t always find the way to stitch the shots. I have had a couple of times when it couldn’t work out the alignment. When I tried the same shots in Photoshop, they worked just fine. Not sure why these didn’t work but the algorithms must have some limitations. Next, it doesn’t always deal well with curvy edges. I have had a couple of stitches that I tried where the aircraft fuselage, although a smooth curve, ended up with some kinks at the area where the stitch took place. Photoshop never caused me trouble with these either. Big panos also seem to make it unhappy with some very odd alignments being chosen (after a long time processing) so they may have to stay in Photoshop for now.

The last problem is cropping. You have the option to have the whole stitch or to have it crop in automatically (or you could crop manually afterwards). Cropping pulls you in to the shot more aggressively to get rid of any blank pixels. You can fix this by shooting a lot wider to have more to work with. However, having been used to being able to fill odd little gaps using Content-Aware Fill in Photoshop, not being able to do so in Lightroom is a limitation. I can, of course, open the file in Photoshop and do just that. However, if I do so, I might as well use Photoshop to do the merge in the first place.

Overall, it is pretty good. I suspect there will be some tweaks behind the scenes as CC gets updated progressively so I might not even know that Adobe have fixed some of the issues. The Fill issue will be more obvious though. I shall probably keep using it unless I feel the Photoshop is merited and it is a good addition but I hope they take it further so I don’t have to consider Photoshop in the future. We shall see because the new update of Photoshop is out and includes Content Aware Fill of the gaps in panos which might be enough to sway me back.

Time Lapse Tutorial

1 Reply

I have mentioned in the past some of my interest in time lapse. I recently posted a sequence on here and a friend of mine asked me to put together a guide to how I had gone about doing it so here you go.

When putting together a time lapse, the first thing is trying to establish what will be special about what you are shooting a sequence of that will make a time lapse more interesting than any other form of presentation. Usually this involves something that changes slowly in real time but becomes more dynamic in a time lapse such as a sunset or storm cloud development. Alternatively, it is something where a lot happens over a long time that isn’t terribly dramatic on its own but, when speeded up, becomes far more impressive. This could be a sequence of activities with multiple people or launching airplanes as was the case for the one I recently made.

Once you have established what it is that makes it interesting, then you have to decide how best to portray that. Choosing your location and your angle of view are the next challenges. I was shooting departing and arriving aircraft at SFO so I needed an angle that gave me a good view of them and a field of view that allowed plenty of the movement to be seen without everything being too far away to be noticeable. In this case, shooting at night works well because the lights stand out far more making the aircraft more conspicuous than would be the case during the day.

Now it is time to get into the technical issues. How long a clip do I want to make? Shooting time lapses means getting a large number of still shots which will each be a single frame of video. Video frame rates vary but I used 30fps for the sequences I make. Therefore, 30 shots will give me a second of video. If I want a minute of video, I need to shoot 1,800 shots. That’s a lot of shots. Also, figure on shooting more before and after the main action since it is nice to have some spare video at each end to play with in future editing.

You know how many shots you need but how long in real time do you want to cover? Are you starting before sunset and finishing well after sunset? Do you have a peak period of activity that you want to cover? Now you can see how many shots you need over a given amount of time so you will work out how long the interval between each shot can be? The interval between shots is critical but you must also factor in how long a shutter speed you want. This is where shooting at night is a lot nicer because you can get nice long shutter speeds. These make for smoother looking videos because there is no jerkiness. Shooting on a sunny day with a very short shutter speed and a few second interval will result in a jerkier output. The “rule” is to have the exposure time at least half the interval. Doing that during the day may require some neutral density filters. At night it is a lot easier to manage.

Now to mount and control the camera. Obviously a tripod is a necessity to make a stable platform for the shots. A cable release is a must have and preferably one with a timer built in so you can program the intervals. However, if you are shooting at night, you can get away with putting the camera on burst mode with a shutter speed to match your interval and then lock the shutter release open. It will then just keep shooting. The other camera thing to consider is image stabilization if you have long shutter speeds. IS can wander around for long exposures making everything blurred. Keep it turned off. I would also switch autofocus off once you are happy everything is sharp to avoid the focus getting changed by the camera. If the conditions are changeable, you might go with something like aperture priority to accommodate changing exposures. At night, manual might be your best bet. Always keep an eye on how the exposures are doing if in a programmed mode to ensure you aren’t exceeding your interval with your shutter speed. There are complex bulb ramping tools available to use if you want to get advanced. I haven’t tried these since they haven’t been necessary for my purposes so I can’t give good advice. I do have the functionality in a cable release device that connects to my iPhone called Triggertrap but I haven’t ever pushed it to its limits.

One thing you will discover is that fixed apertures are not fixed. Cameras will go to a slightly different aperture for each shot which can result in slight variations in exposure. This “flicker” can be managed by clever manipulation of the lens position but I prefer to use software to fix it. The software will also help if you end up tweaking the camera settings during the capture sequence.

Last piece of equipment advice is bring a chair. Once everything is at work, you don’t have much to do. You can shoot with a second body of course or you can sit back and relax. If you have a buddy along, that is not a bad thing! I shoot all of my sequences in RAW format. For night shoots this gives you a lot of latitude for tweaking the exposures. It does use up a lot of memory but storage is cheap these days. With charged batteries and big cards, you are good to go.

Once the shoot is done, you get home with a ton of images that look remarkably similar. First I import them into Lightroom. I will keyword them in the same way as anything else from a shoot including adding a time lapse keyword. Then I will make sure all of the time lapse shots are in one subfolder before jumping over the LRTimelapse. I am not going to try and write a tutorial on using LRTimelapse. The website for the software has far better guides to how to use everything. However, I will focus on the key elements I go for.

The main one is the Deflicker process. This is a routine that analyses each shot and applies little tweaks to the exposure to avoid any visible flicker. You can draw out a box in the image preview to tell it where to analyze. This means you can pick an area that is constant in the image as a reference and avoid areas where there is a lot going on. If the image exposure is gradually changing, it won’t affect that. It will just take out the individual variance. You can see a trace of how the overall exposure tracks during the sequence.

The other element I sometimes use is panning. You can define keyframes in the sequence and jump back into Lightroom to crop each one as you wish. Back to LRTimelapse and it will calculate the individual crops for each frame to make a smooth pan between your chosen keyframes. Quite nice to provide a little variety if you wish. I always crop the images to a 16×9 format since that works nicely for HD output formats.

When all of your tweaks are fixed, you save the metadata files and go back to Lightroom to reload them. (There can be a bit of back and forth like this throughout the process.) In Lightroom, there is an export dialog from LRTimelapse which then renders out the individual files from each shot. These will then be taken by LRTimelapse and rendered as a video clip.

Sounds really simple that way and, to be honest, it is pretty straightforward. A little practice helps of course. When it works out, the result can be very satisfying. Of course, a lot of times, you see something that you really wish you had done differently. That is something you will have to put down to experience and try to remember next time you are out shooting.

One last thought – if you are shooting something that involves light trails at night, consider making each image a layer in Photoshop and blending them together as a stack to see what happens. It can work pretty nicely sometimes. However, you can end up with a pretty huge Photoshop file so your computer may groan while processing it. Happy shooting!

HDR 32 Bit and Lightroom

Leave a Reply

This post is a plea for help. Anyone who is a regular user of Lightroom and Photoshop may be in a position to educate me a little. I use Photoshop to process my HDR shots. I start out in Lightroom, select the shots and use the Edit Photo>HDR Pro method to open them up in Photoshop. I then use the 32 bit version of the HDR Pro processor to create the file. I then take the Edit Using Adobe Camera Raw option for opening the file to undertake the final mapping. This works pretty well and I can usually get something I am happy with.

I then close the file and save it which automatically reimports the finished file back into Lightroom. This is when things go wrong. The view of the file in Lightroom doesn’t appear to make use of any of the edits I had done in ACR. The highlights are too bright and the shadows too dark. I can then use the Lightroom Develop settings to get something close to what I want but surely that should not be necessary. I might make further tweaks but shouldn’t it look the same when I close out of Photoshop.

Below is a sequence of screen captures to show you the unprocessed shot in ACR, the finished version, what it looks like in Lightroom and then what I can tweak it to. Any suggestions are gratefully received. (Note, I didn’t make this exactly match the two edited versions. It is aimed to illustrate the disconnect and the recovery process.)

Original HDR merge pre tone mapping

HDR with ACR tone mapping applied

How the saved image appears when imported to Lightroom

Using Lightroom to get back to something like I had in Photoshop

Negative Scanning

Leave a Reply

I have a lot of negatives and transparencies from my days of shooting film and various of them have been scanned at times either for projects or just because I wanted to have them as part of my digital library. For a long time, I have made use of a Minolta Scan Dual III film scanner. While it doesn’t appear to like modern operating systems, a few tweaks to the config files make it run with Windows 7. However, it is not a totally smooth process and the scanner has a habit of hanging at odd times which requires a lot of fiddling to recover.

When I built my latest computer, I didn’t bother to install the drivers and the box is sitting on a shelf. I was figuring I would come back to it if I needed to. Recently, I needed to scan a single strip of film and wondered whether I had an alternative. I do have a flatbed scanner that has a film scanning capability. I had previously dismissed flatbeds for film scanning since they were seen to be inferior to dedicated film scanners. However, my film scanner was pretty old and gave me mixed results and my flatbed is relatively new. Therefore, I figured I ought to give it a go.

First, the limitations. It will scan a strip of film of shots but you have to move a small device between each shot between scans. No scanning a full holder of film or slides. Big scanning jobs are therefore a lot less practical. As for the image quality, I have yet to find a way to have much control. A preview scan is done but if it comes out with a poor exposure, for example, I have yet to find a way to adjust it. The dedicated film scanner allows all sorts of tweaking. Then there is the speed. If I go for the highest resolution, the scans take many minutes for each exposure. I can wander off and do something else but, if you have more than a couple of images to scan, this is a slow and inefficient process

As for the output quality, it is a mixed bag. The scans are not too bad. Despite the lack of control, if the original is a straightforward image, it comes out okay. Large areas of a similar color such as a sky are subject to banding. I suspect this is a problem all of the time but most pronounced on large areas of one color.

So, am I going to make do or is the film scanner going to be resurrected? I think I will get it back off the shelf. The time and quality issues are sufficient to make the flatbed not helpful except for odd individual scans. It does a nice job on documents but that is it. Now I shall have to find a location for the film scanner and sort out the config files again.

Adobe Camera Raw and Video

Leave a Reply

Regular readers will be familiar with my gradual experimentation with video and video editing. In the early days of my playing with video, I discovered just how sensitive video is to exposure errors. The sort of thing that could be easily corrected in post on a raw still file were not so easily dealt with in video that was already compressed when it came out of the camera. Avoiding over exposure was one thing to do which is slightly different to shooting aircraft against a bright sky when having enough shadow detail on the aircraft is important and the sky can be recovered a bit to give a more pleasing outcome. For video this doesn’t work as well but having less shadow detail doesn’t seem to matter with motion as much as it does for stills.

Another change I made was to go with a development profile in the camera that is a lot “flatter”. This was something I read about on various blogs in that it gives the editor more to work with when grading the video later. This is certainly true but it means you definitely have to do some work in post to get the image back to something more pleasing.

I never cared for the editing tools in my old video editing application since they were weak and not as intuitive as what I was used to in Photoshop/Lightroom. Consequently, I embraced video editing in Photoshop when it became a better developed feature. However, I hadn’t found it as easy as I had hoped to get the right effect using levels and curves adjustments. Then it struck me (why it took so long when everyone else must have been doing this) that Camera Raw is the tool that combines all the things I need to enhance the video output.

I have now been using it on a couple of projects and I have to say it works very well. You have to convert the layer to a smart object first and then apply Camera Raw as a filter. If you don’t convert it, the filter only applies to the frame you are looking at and that is no use. However, one lesson I have learned is to work out the other parts of your edit first and leave this step to last. The rendering of the video with the filters applied is a huge amount slower. Short videos that would previously have rendered out in a minute or two may now take over an hour. This also applies to running through the video to make your edits. The real time rendering is lost. I suppose that you will be using a proper video editing suite if you do this on a daily basis so the use of Photoshop is moving away from its core role. However, it suits me to do so. Therefore, make all of the edits you need first and when you are happy with the final composition, convert to smart objects and filter away. Just remember that the conversion gets rid of transitions so those will have to be put back in again but that doesn’t cause me any problems. Then hit render and go and do something else. While Photoshop will do many things in the background, video rendering takes it over and you can’t work on another project while it works in the background so go off and write a blog post. That’s what I am doing right now!

Funky Cloud Processing

Leave a Reply

Part of a run back into Oakland recently came across the bay but it was unfortunately covered in cloud. However, the cloud was sitting in very distinct locations with a clear edge as you headed down the bay and a similar edge near the shoreline on the Oakland side. Since we were not high above it, this looked pretty interesting. Even as I photographed it, I figured the shots would need something a bit different when I got around to processing them. Having a more contrasty look seemed the best bet and a black and white conversion also seemed likely. That is what I went with and I was quite pleased with how it turned out. See what you think.

HDR Panoramas

Leave a Reply

Another one of my processing technique posts today. For those of you interested in pictures of places, today will probably be one you pass on. You have been warned. This is about my first venture into the realm of HDR panoramas. I know at least one of you who knows exactly how to do this sort of thing and does it on a regular basis. You also will probably skip the rest of the post. However, you have some specialized tools for doing the job and I am playing with Lightroom and Photoshop so here is how it goes.

My initial thought having taken the shots was which order to carry out the processing. HDR first or pano first. I concluded that it had to be pano first. All of the pano exposures were consistent and would stitch properly while I wasn’t convinced that each of the pano frames would be consistent if I had done the HDR blending first. However, this left me with a second concern. Would the pano merge produce images that would align for the HDR merge. I use the pano tools built into Photoshop and, while I select the algorithm it uses, I did not have confidence that it would produce an identical alignment for each set of exposures. However, this was the route I tried.

Stitching the panos was straightforward enough. I created each of them from Lightroom and ended up with five panos with differing exposures. At this point I could have taken them directly to HDR Pro within Photoshop but, since I wanted everything to end up in the Lightroom catalog, I decided to save the files and go to HDR Pro from there.

Here I encountered my first hiccup. As expected, the panos produced were not identical. There were very close but not identical. HDR Pro only works on files that are the same dimensions. I imagine some more specialized HDR applications might be able to handle this but I was stuck with Photoshop. Since the panos were thousands of pixels across and only a few pixels different, I opened them back up in Photoshop and changed the canvas size to be identical in each case. HDR Pro is able to manage alignment of slightly misaligned shots anyway so I wanted worried about the positioning. Also, with such small changes in dimensions, I didn’t fear that I would have distortion.

With this change made, Photoshop went to work and created the HDR file. Amazingly, it worked just fine. I didn’t have any problems with the files being distorted relative to each other and it did a great job of blending them. All that was left was to crop everything in to clear up the empty corners from the pano creation (I didn’t get rid of those in the first instance since I was trying to keep the pano files identical in size and alignment) and then a few tweaks back in Lightroom had the job finished. I was pretty pleased with how it worked and, with the experience of this time, should be able to turn them around quite quickly next time.

Boeing 777-300ER Main Gear

Leave a Reply

When Boeing launched the 777-300ER, they took the stretched fuselage of the 777-300, a model that didn’t sell particularly well and married it to the updated wing that made use of the fuel capacity of the outboard portion of the wing that had been left when the original concept of a folding wing was contemplated. The increased the weights of the jet, added far more powerful engines and, with the increased fuel capacity, came up with a winning formula that has done a very effective job of killing off the 747.

One problem that they had to deal with during development was runway length requirements for takeoff. Even with the bigger engines, the long fuselage limited rotation angles at takeoff and meant a higher takeoff speed was required which meant a longer runway requirement. Boeing came up with an interesting solution (after dumping some slightly more curious ideas). The main gear on the 777 has a triple axle bogie. Previously this had rotated about the pin attaching it to the main gear leg. Boeing’s solution was to lock the bogie level during takeoff.

The result of this is to have the rotation of the jet at takeoff to take place around the rear wheels of the bogie rather than the gear leg pin. The slight aft movement of the rotation point allows the aircraft to rotate slightly more nose up and gain a greater angle of attack. This gives slightly more lift for a given speed. This means an earlier takeoff and a shorter runway requirement.

I have tried many times to witness this at work. First, it happens pretty quickly. Second, I am often in a poor position to see the rotation point. Recently I was at SFO to pick up some people. I was getting a few shots prior to their flight arriving and a Singapore 777-300ER was taking off. The rotation point is quite far away (although, if you are in the terminal, you might have a good view) and the heat haze is a problem. However, I decided to get a sequence of shots anyway. Now, how to use them.

Heat haze is crappy on stills but less of an issue with moving images so I decided to animate the sequence. I imported all of the shots into Photoshop as layers in a single document via Lightroom. The hardest part was aligning them. I started at the bottom layer and then progressively made each layer above visible. I then changed the latest top layer blend mode to difference. This makes aligning them a lot easier since everything is black unless it is different. I was focused on the gear so used that as the reference as the fuselage rotated. Once each layer was in place, I changed the blend mode back to normal and moved to the next layer up.

Once they were all aligned, I used the animation timeline to make frames from each layer (and reversed the order since every time I do this they seem to be the wrong way around). Then I could crop in to get the overall view I was after and save the file. A Save for Web allows the generation of the animated GIF and we are done. The image at the top is the final result. It does allow you to see a bit of what is going on if you look closely although it is still a bit hard given the distance, the angle to the ground and the heat haze. I guess I will have to find a location closer next time.

Updated Approach to Lightroom Catalogs

Leave a Reply

I have been a user of Lightroom since Adobe release version 1 quite a few years ago. It has been a very useful tool for me and something that dominates my workflow. In the early days, there were issues with the number of images that a catalog could have before it started to respond sluggishly. Consequently, I created a series of different catalogs for different subject areas. Military aircraft were in one, civil in another, wildlife in a third, sports in a fourth and so on. This system was fine to use although it had a few minor drawbacks in that some images would be in more the one catalog. I would import them to one and then export them to the other. However, as I edited one version, the other one did not always keep up despite saving the sidecar files to disc.

At some point, Adobe updated the functionality of the application and it was no longer so constrained by the number of files in a catalog. However, I had a series of catalogs that I was familiar with so I didn’t pay much attention to this change. However, over time, it occurred to me that I was making my life more complex than it needed to be. I could probably cut down the number of catalogs dramatically and make the workflow a bit simpler.

I decided to have one catalog for all of my aviation related imagery and the other for everything else. To do so, I created a new catalog for each of these and started importing from the other catalogs. This was not as smooth a process as you might have imagined. There were duplicate images as I knew and you could set the system up to make these virtual copies so nothing got lost along the way. However, sometimes the import did not go well. They were large catalogs coming into an even larger catalog and this caused some struggles. I had to delete and start again at some points but ultimately I got it to work. I did have to recreate some of the collections which did not transfer so easily in some cases but it is now done.

Did it make life easier? Yes, it did. Having just two catalogs is now a lot more straightforward. Do they run just as well? No. The aviation catalog does seem to be a bit sluggish sometimes. Usually it works fine but it is definitely not as responsive as the individual ones were. I shall see how this develops over time. A new machine is in the works so whether that will make a difference or not, we shall see.

Playing with Blending Layers

Leave a Reply

I have been making some shots with multiple exposures to overlay. This is something I have posted about before and the shots here are similar to those from before. However, this post is less about the shots and more about the post processing I used. Previously I opened up all of the shots as layers in a single file and then auto-aligned them. Once done, I then used the Auto Blend functionality to show each shot o the aircraft in place.

This was a lot quicker than my previous approach and was something I picked up from posts on photographing star trails. However, recently, I have not been as happy with the results as I should have been. Some of the planes, particularly those near to the edges, had some odd artifacts appearing. Also, if there were any overlaps, the blending masks could give some weird effects. Therefore, I have taken a different approach for a while. This is slower, I admit, but I think it gives a better result.

Once the alignment of the images is done, I hide them all except the bottom layer by Alt clicking on the eye beside the last layer. Then I add the next layer up back in but mask it out completely. A white brush on the mask then allows me to paint back in the new aircraft positions. This is a bit laborious but it does allow you to decide exactly what you want in and what you don’t. if one file is not helpful to the composition, you can easily ignore it.

If the layers are not all exactly aligned from shooting on a tripod, you will also get gaps at the edges on different layers. You can also fill these in by brushing in the layers that provide the right coverage and get a complete image. Once you are happy, flatten the whole thing and you are done.