The Roving Photographer

Blog

How I Shoot Panoramas

I got a request from one of my readers to talk about how I shoot panoramas. I am always thrilled to get feedback from you guys, including requests. So without further ado, here’s part one of a two-part article that will hopefully show how I make these work:

First off, I like shooting panoramas. There’s something in the nature of the final images that strongly appeals to me, the sweeping nature of the image and the way that it shows so much more context and sense of place than a single shot is able to. That said, panoramic shooting isn’t necessarily something I plan on. It’s somewhat uncommon that I head into the field with the express intention of creating one, but rather its something that occurs organically once I’m there and see a scene that would be best expressed that way.

Panorama of sunrise at Bullhead Bay, Lackawanna State Park, on the morning of 22 January

{Click to view larger}

There aren’t any hard rules for this. Strong horizontal elements are naturally necessary, but aside from that I shoot panos with more emphasis on the feeling that “I cannot capture this scene with a single image, but feel that it’s worth capturing nonetheless.” It’s a huge advantage of digital that I can shoot multiple frames and merge them with relative ease; it’s an even bigger advantage that I can see the frames as I shoot them, and therefore accurately judge the composition and overlap.

Once I’ve decided to shoot a panorama, I have to decide how I’m going to accomplish it. Shooting from a tripod is generally accepted as being the “proper” way to proceed, but I often fail to bring mine along, especially if I’m planning on covering ground. So there is hope for handheld panoramic shooting as well.

In either case, I always shoot a little wider than the scene I want to capture. This means two things: first, I use a slightly wider zoom setting (ie: I shoot “loose”), and second, I extend the panoramic series father on both the starting and ending sides. The reason for this is simple: when assembled in post, the resulting image will be uneven. Either the ends or the middle will be wider, or sometimes the layout follows a sort of shallow curve. Either way, I’ll have to crop in to achieve a nice, rectangular image, so allowing myself some extra space makes this much easier. (Failure to do this results in some frustrating Photoshop time, trying to clone in missing corners. You quickly learn to shoot loose in the field.)

Panoramic view of Lefferts Pond in Chittenden, VT, on the afternoon of 29 December 2011.

{Click to view larger}

Shooting a pano always begins with locking in the camera settings. I’ll meter different parts of the scene, and usually snap a few test shots just to nail down the exposure, before switching to Manual. If I’m shooting from a tripod, then slow shutter speeds aren’t an issue and I’ll stop the lens down to at least f/11 at ISO 100 and engage the Mirror Lockup and 2-second self timer (or use my remote, if I remembered to throw it in the bag – but really I find the timer to be easier, and one less small accessory to carry around/lose).

If I’m hand holding for the shots, then shutter speed becomes essential, and the faster the better. Nothing ruins a panorama faster than camera shake on a single image in the middle of the series. I may trade aperture for shutter speed and I’ll definitely increase the ISO. I won’t engage high-speed continuous shooting, however, since a consistent composition is more important that spray-and-pray ability.

Tripod mounted, I usually shoot verticals. It takes more shots to cover the whole range, but the final resolution is larger. Hand held, KISS is the rule to follow, and fewer frames means fewer chances to screw up, so I’ll usually shoot horizontals. The number of frames in the series depends entirely on the size of the scene I’m trying to cover, as well as the focal length of my lens. What matters most is the overlap between frames. The more overlap you include, the easier it is for the software to match features and create an accurate panorama. I try to shoot each frame with a 33% overlap. I always work left to right, so the right third of my first frame and the left third of my second frame contain the same portion of the scene, and so on.

{Click to view larger}

Realize, however, that overlap is good, but there is such thing as overkill. I find approximately 33% to be a rule of thumb that results in good panos. If I’m shooting handheld, or in a hurry (fleeting light/subject, whatever), I may only use 25% overlap. And on the other end of the spectrum, you have to remember that the more overlap you shoot, the more frames you’ll have to merge. More frames = increased demand on your computer’s resources, and lower-end systems may not be able to handle it. (More on this in Part 2.)

Here’s the catch! If you read any articles/websites/books about shooting panoramas, you’ll certainly hear mention of nodal points, followed by some sort of sliding tripod attachment meant to properly align your lens. The short version is this: to make an accurate, distortion-free panorama, the point of rotation shouldn’t be the camera itself (which it is, using a standard base plate and tripod) but the nodal point in the lens. Achieving this means that the camera needs to be moved backwards until the nodal point is directly over the tripod’s point of rotation. Hence the slider accessories and special tripod heads. (Some good news: if you have a telephoto lens with a tripod ring, it will be much closer to the nodal point than the center of the camera itself. Bad news: you’d have to shoot a panorama with a telephoto.)

Personally, I don’t worry about nodal points. I shoot my panoramas on a normal tripod and accept that there will be distortion. Either I work with it in Photoshop (cropping, free transformations) to fix it, I accept it as part of the artistic nature of the image, or I delete it. If I get really serious about this in the future, maybe I’ll invest in the equipment. Or if I get inspired someday, I’ll DIY a nodal slider. But for now, I shoot panos because I enjoy it, so I’m not going to stress over it.

{Click to view larger}

A few final thoughts on shooting a pano. I’ve already said that I lock the settings in via Manual mode in the camera. This is essential – you don’t want things to change from one shot to the next, or they will not match up properly. This goes for lens filters, too. Used across a sweeping range, a polarizer will show an ugly gradient. I’ve tried it, I’ve deleted the resulting files. Don’t bother. A graduated ND filter can be used with in a scene with a level horizon, but again, you need to evaluate what you’re shooting and make sure that the effect will be consistent across the entire range.

I lock my focus in manually as well, for this same reason. No fancy method here, I simple visualize where the focus should be, and then pick a representative spot to auto-focus on, some feature that is at a distance that is fairly consistent throughout the range. Once focused, I switch the lens to manual focus, align for my first composition, and start shooting.

During the composition phase, I make an effort to level the camera to the horizon in the center of my capture range. As I rotate the camera, it may go off-level to either end of the range, especially combined with distortions, and this is something I just have to accept. But I want it level in the center to start with.

As for actually rotating the camera, my tripod has a rotating center column, and I just spin that around as I go (making sure it’s plenty loose before I start). Some tripod heads have an isolated panning adjustment – that’s good, too. However you do it, you’ll want the tripod securely placed and smooth, easy rotation. Or, if you’re hand-holding, you want to be in a position where you can keep the camera to your eye and turn in place without falling over or into anything. Common sense, there…

The last thing I’ll do after shooting a pano is shoot a junk frame – something intentionally blurry, or with my hand in the image, just so I know that the pano series ends there. It’s one of those simple things that can make life easier on the computer.

And that’s it – there’s nothing terribly scientific about my process, but it does seem to work. What I like most about it is that it’s quick. I don’t spend a lot of time futzing around with extra gear or making measurements or any nonsense. I get in, I setup, and I make the image.

Check back on Saturday for Part 2, which will cover my post production!

{Click to view larger}

Brent Pennington

Related Posts:


Leave a Reply

You can use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

RSS Feed

Categories

Fine Art Prints Available

Sure, my photos look good here on the site - but they look outright amazing when printed big and hanging on a wall! Check out my fine art gallery, with a selection of prints available in multiple sizes, on canvas and even metal, and all printed by a professional lab.


TRP | Fine Art Prints Gallery