The new lens correction tools in Lightroom 3 bring perspective correction to the RAW workflow. Previously, I’ve had to go into Photoshop to make these corrections, but now they can be done right inside of Lightroom, which saves a few steps. However, I still prefer the Photoshop controls over those in Lightroom. I’ll try and explain why…

The image at the top of this post is the image after correction in Lightroom. Here’s the original image:

You’ll notice that the windows are tilting in toward the center of the frame — it’s perhaps most noticeable on the far right side where the edge of the building is leaning in as well. I take a lot of these kinds of shots and do my best in the field to keep everything straight in-camera, but sometimes you’re in a tight spot. Shooting wide and looking slightly up… you’ll get this kind of distracting distortion. A tilt-shift lens or a camera with bellows would take care of the problem, but you don’t always have that option available so it’s nice to be able to solve this issue via software.

In Lightroom, you have a pair of sliders for vertical and horizontal correction. Moving these sliders creates an effect similar to tilting a print in your hands side-to-side or up-and-down. If you ever controlled perspective shifts like this in the darkroom by tilting your easel, you’ll be familiar with the effect. It works reasonably well but it’s a little like a funhouse mirror and any large movements introduce some strange, new distortions. In the end, you have something that looks like this:

The vertical and horizontal lines are now straight (except for the sidewalk which, in the case, actually did run downhill like that) but you’ll notice that we now have grey “background” showing on the bottoms and sides of the images. Lightroom makes its perspective corrections inside the frame so you end up with an image that is smaller than the canvas size, to use Photoshop terms. You need to crop the image down to get rid of these grey wedges. Not difficult, but another step.

I’ll use this method a lot I’m sure. It’s really convenient to have this ability inside Lightroom. However, my preferred method of perspective correction is still part of Photoshop: the Distort control under Edit > Transform.

I turn on the grid in Photoshop first, to better see where things are skewed. Then, Select All and you’ll get corner handles on your image. It’s best to be zoomed out so that you can see some free space around the border of your image. By clicking and dragging these corner handles, you can stretch the image into alignment with the guides:

You can see the top-right handle has been pulled beyond the image frame here, eliminating the left-tilt of the right side of the building. The image will not show beyond the edge of the frame but the data is still there if you drag the corner point back. This gives you a live preview of what the final, cropped image will look like. Better yet, we’re making this correction to the outside of the frame so that there is no need to crop the image afterward. The perspective correction and crop are being handled simultaneously and there is no chance of an edge artifact getting left in your final file.

There is a Perspective control in the Transform group as well, but I generally prefer Distort. The reason is that Perspective tilts the image plane like Lightroom does — symmetrically. If I tug the right side out, the left side will move correspondingly. With Distort, each point moves individually — move the top right handle and only the top right corner distorts. I find this much more useful because the corrections I need to make are not always symmetrical.

Lightroom has a good set of correction tools — there are rotate and pincushion corrections there as well — but I’m glad to have the ability to go to Photoshop to fine tune images when needed. If I can do 90% of the image correction I need in Lightroom, I’m happy to jump over to Photoshop for the last 10%. No tool can do everything. When they try, they get incredibly bloated. Honestly, I kind of hope that Lightroom never adopts layers, etc. Keep it simple and I’ll go to Photoshop when necessary. The two programs work very well together.

I’ve been giving Lightroom 3 (now 3.2) a real workout for the last few weeks. I’ve been importing my back catalog in massive chunks and making a quick pass at adjusting the original RAW files rather than importing the images that I had already tweaked previously in Adobe Camera Raw. This is a pretty big-stroke process right now — I’ll come back later and expand keywords and tweak the highly ranked shots — but it’s been a really interesting process to see the images improve. I’d like to think that the improvement is due to my own advancing skills in RAW processing, but the fact is that much of it is really due to improvements to the software over the years. Better processing algorithms and better controls.

I’ve been importing my 2006 images from Chile most recently and have noticed a couple of major advancements. One is Lightroom’s much improved ability to get rid of noise in high-ISO shots. Images that have never seen the light of day will now be going into my stock files because the noise that would never have passed Quality Control is now all but gone. Amazing. I’ll share some examples in the coming posts.

The most noticeable change in the images from Chile, though — especially Valparaiso — is how much more control I have over color now. Valparaiso is a city that is all about color. Colors you never thought you would see together, you’ll see in abundance in Valparaiso. Capturing some of these colors on film would have been frustrating. Sure, Velvia would have made that row of multi-colored row houses really pop. But then you’d round the corner to find a weathered, grey facade with the most subtle hues and you’d be cursing yourself for having 20 more exposures of that super-saturated film in your camera. One of my most favorite things about digital is just that — you can make decisions on a shot-by-shot basis and not be stuck with a film decision you made 10 exposures ago. Did you pop in a roll of ISO 800 film to shoot that dark church interior? Too bad now that you’re out in full sun and all kinds of contrast. No, being able to adjust ISOs and saturation levels for each individual image is a game-changer.

And that flexibility doesn’t stop with in-camera controls. Lightroom (as well as all other image processors, really) lets you tweak color and contrast in ways that film selection alone never even came close to. The Vibrance slider in the Presence panel has become one of my favorite tools. More subtle than the Saturation slider, it affects the blues and greens more than already-saturated reds and yellows. Very nice. And the sliders for adjusting the hue, brightness and saturation levels of individual colors take that control to another level.

One other set of controls that I find myself using more and more often is the new perspective correction panel — actually the “transform” controls in the lens correction panel. While I still prefer the “distort” controls in Photoshop, it is nice having some of this ability in Lightroom. The controls take some getting used to but they are a welcome addition.

I’ll go into more detail on each of these and other Lightroom features in future posts and show some examples.

I returned to the Eldridge Hotel for another twilight shoot last night. You may remember a recent post where I shot it from a rooftop across this street. This time I went street-level and parked myself under a tree along the sidewalk catty-corner from the hotel. I was again going for a long exposure with some traffic streaking and a combination of the dying light in the sky and the warm street lights.

I’m discovering that this particular building is not lit very well, nor do there seem to be many lights on in the windows in the early evening. My solution has been to merge several shots together. The image above was a quick attempt this morning to see how it might look, but I’ll need to do a more involved layering later. While I was locked down on a tripod, I was also changing apertures. The differing depth of field between shots means that I need to mask out some areas to preserve the focus where I want it. Otherwise I end up with ghostly halos around the foreground elements.

The advantage I gain by using multiple shots is that I can use one as the base image — here I’ve chosen a later shot with a good sky — and then “paint in” elements from other shots. One example is using a shot taken earlier in the evening where the building itself had more light on it. While it’s nearly silhouetted in the base image, I put the earlier exposure on a layer above, added a black layer mask to hide it entirely, and painted the building back in using white as my foreground color. I could reduce the opacity of the upper layer to better blend it and achieve the level of light that I wanted on the building.

Another advantage was that I could get more traffic streaks than what actually appeared in any one photo. I picked dark shots from my bracketing that had good headlight and taillight streaks, stacked them on layers above the base image and set those layers to “lighten” in the layer mode menu. That made only the areas that were lighter than the base image visible — in this case, the light streaks.

Headlights from the street to my left would also occasionally throw some light into the leaves of the tree above me. I was also able to paint in some detail there using the same technique as I used on the light streaks.

The downside of this kind of shoot is that I’m tied to one position for about an hour. I usually move around and find different angles, trying to make the most of the light as it changes and fades. Here, I need to dedicate myself to one angle and stick with it in order to have all the layers I need for the final product. Last night the sky was surprisingly dramatic for a short period in the middle of this sequence an I regretted positioning that tree where it hid the sky but I was committed to stay put in order to keep all of the images registered. Since this is a local shot for me, it’s not such a bad thing. I’ll just keep going back and trying different things.

It’s ridiculously hot out there — heat indexes well over 100 for days now — but the purple coneflowers in the front yard don’t seem to mind. I’m hoping it eases up a little tomorrow evening so that I can make another attempt at a twilight shot of the Eldridge Hotel. Finger’s crossed…

I’ve been meaning to write this post for a long time. While finishing up the keywording of my Thailand files in the last couple of weeks, I ran across a good example of what I’ve been wanting to discuss, and that is: how I use Google Maps to help me identify my images.

Of course, taking notes on site is always the best practice. I carry a small notebook in my camera bag but, in truth, I tend to take most of my notes with the camera itself. Here’s an example: on my last day in Chiang Mai, I had a few hours in the morning to stroll the old city on my own and photograph the various Wats, Chedis and Prayer Halls. I had heard that Chiang Mai has over 1000 Wats so, suffice it to say, it’s important to keep track of which is which when it comes time to keyword and label for my stock agencies.

I set out from the hotel before dawn and headed for the original, walled city. Once inside the walls, I would pass a small temple complex nearly every half a block. While I had been driven through these streets regularly for three days, I never would have seen many of these buildings without being on foot. They are tucked away and need to be sought out.

When I would begin to shoot a temple site, I would first look for a sign with the name in english and photograph that first. This is my way of taking notes. The images can always be sorted by capture time and I’ll have an “opening title” for each location I visited. Hopefully it will even be spelled correctly, but, you never know.

When I arrived at the Wat at the top of this post, I couldn’t find a sign in english anywhere. Nor was there anyone around that could tell me what it was save for a group of monks (who I opted not to bother), chanting inside this main building:

The site was a monastery comprised of a viharn (the building in the foreground), a chedi (the gold spire behind it) and a ubosot (the roof seen peeking over the trees in the back on the right). The english terms for these might be an assembly hall, a reliquary, and a prayer room or ordination hall. The ubosot, or prayer room, is what is pictured in the opening photo.

As it turned out, I never had the opportunity to speak with my local guide again and I continued on my way to Bangkok. Once back in the office at home, I could have tried to contact him by email but decided to do some detective work on my own first. This is where Google Maps comes in.

I knew the route that I had walked through the walled city, and I knew from signage, what the names were of the Wats on either side were. I zoomed in close in this area with Google Maps and switched to the satellite image view. There was my Wat:

You can see the shiny gold chedi in between the two red-roofed buildings just above and to the left of center in this screen capture. There is a label, but it’s in Thai. I tried using an online translator but didn’t trust it so I tried using Google’s street level viewer in Google Maps. That’s the tool shaped like a little man. While the area had not been mapped for street level views, the tool did show several blue dots for the positions of still images that were on file for the site. Sure enough, by looking at some of these images, I was able to confirm that this was the Wat I was looking for and that it was named Wat Chai Phra Kiat, or Monastery of the Renowned Victory.

That’s the image that convinced me. Still, just to be safe, I Googled the name and found several other sites that confirmed it was what I was looking for and provided me with some alternate spellings.

The internet has been one of my most valuable tools for researching images and compiling keywords. I would hate to imagine the time it would take to track sites like this down through books, emails to local guides, etc. I still do my best to identify sites while I’m on location — there’s no substitute for that — but at least in situations like this, I can feel confident that I can come up with the information later when necessary.