A neighbor of mine asked me this weekend to stitch together a set of images he took in New Zealand into a panorama. It was an easy enough task, except that it seems the camera was on auto when the photos were taken. This led to very nicely exposed individual images, but there were exposure differences between the shots. With exposure the differences, the images in your pano might stitch well, but it will be obvious where one ends and the other starts. In this case, it took some Lightroom and Photoshop time to get the images tweaked so they better matched.
I’ve taken a few panoramic images (four of which are shown here) and along the way have learned a few secrets to successfully shooting panoramas. The two main secrets to shooting successful panoramas are to 1) keep all the camera settings the same for all the shots – this includes exposure, focus, and focal length, and 2) to move the camera along a level plane, typically either horizontal or vertical. Here are some hints to shooting panos:
- shoot in manual exposure mode – use your camera’s meter to get a f-stop and shutter speed, than turn the camera to manual mode and set the f-stop and shutter speed to the same settings; do not change them through the series of shots
- for jpegs, shoot with manual white balance – do not use auto white balance, pick one setting (such as cloudy) and leave it there
- better yet, shoot in raw – and then process the images exactly the same way prior to blending (more on my workflow below)
- shoot in manual focus mode – use the autofocus to set the focus, than turn it off and shoot all the images without changing the focus
- use a tripod – to help keep the camera level and moving in a single plane; if you don’t have a tripod, be careful to move the camera in a single plane. When handholding, most people have a tendency to sweep upward or downward. Even with a tripod, without special equipment, it is difficult to a good series of shots without some movement off your preferred plane.
- consider photographing with the camera vertical for horizontal panos and horizontal for vertical panos – though following this advice will result in more images , it will give your panorama more width (and more fudge room for imperfect sweeps)
- don’t compose the main subject too close to the edge of the frame – after stitching the images together, you will need to crop off where the frames do not line up exactly; you don’t want to crop off part of your main subject
- it’s better not to use a wide-angle lens – wide angle shots have distortions which make it more difficult to stitch properly
- compose the first shot at either end of the pano, then take a picture with your hand or fingers in front of the lens; do this again at the end – this marks the beginning and ending of the series, making it easier to figure out which images belong together when doing the stitching
- overlap the shots by at least 20 to 25% – I typically look for some distinct feature about 1/3 off the right side of the frame (when shooting a horizontal pano sweeping rightward), take the shot, than recompose with that distinct feature on the left-hand frame edge for the next shot
- shoot fairly quickly – to avoid having changes in light, clouds, etc. between frames
I almost always shoot in raw, and my basic processing workflow for panoramas goes like this:
- Import the images into Lightroom and adjust the white balance (even if the auto white balance looks fine, move the sliders a little so auto is no longer selected) and correct the chromic aberration on one image.
- Copy those adjustments and paste to all the other images.
- Select all the images, right click on the mouse, and select the merge in Photoshop option.
- Allow Photoshop to merge with its auto settings – most times this works well, occasionally I’ll need to try different setting or even do it manually
- After Photoshop merges the images, check the seams to see if they match well, and if so, save the file and go back to Lightroom.
- In Lightroom, select the Photoshop file just created, go to the Develop module, and now start my normal processing workflow (which is the subject of another post)
Occasionally friends ask me how to take better photographs. Often the question centers around whether to buy a new camera. Will buying the newest DSLR or point-and-shoot give them better pictures? The answer is generally no. For novices, the simplest way to get better pictures is to use good composition when taking photos. Better equipment will only help if you have the basics of composition down to begin with. Without good compositional skills, a better camera will just give you higher resolution bad images. You might be able to make a 20×30-inch print without sacrificing detail, but the detail will still be boring to look at. To a professional, nothing screams “amateurish” more than placing the subject dead center in the frame. (Of course, like all rules, this one occasional can sometimes be broken with good effect, but trust me, until you know better, don’t break this compositional rule!)
While whole books have been written on composition, the easiest compositional trick to better looking photos is to avoid the “bull’s eye syndrome.” A corollary is to use the “rule of thirds.”
What do I mean by the bull’s eye syndrome? As you might know (especially if you are a dart player like I am – hard tips only please) the bull’s eye is the center of the dart board. A photo with bull’s eye syndrome has the subject of the photograph dead center in the frame. Nothing could be more boring! Imagine you are taking a picture of your sweetie. If you place her (or him) dead center in the frame, making a bull’s eye of them, you leave a lot of dead space above their head. If you are trying for a full body shot, by placing their face in the middle, they are twice as small in the frame as they would be their face was placed at the top of the frame. If you remember one compositional rule, make it this – do not place your subject in the center of the frame!
So, if not in the center, where do you place a photographic subject, you ask? That’s were the rule of thirds comes in. Imagine your frame divided into thirds both horizontally and vertically, making a grid. Some point-and-shoot cameras even come with an option to show these lines on the LCD screen when photographing. If your camera has this option, be sure to use it! Now, whether you can see the lines or just have to imagine them, place your subject on the intersection point of two of these lines.
You can also use the lines instead of just the intersection points. If a horizon is in the scene, place it on one of the horizontal lines. If there is a tall vertical object in the scene, a tree or tall building perhaps, place it on one of the vertical lines.
The photo above illustrates this principle. It is a photograph of my wife, Tanya, that I took down on the beach in Oregon a couple of years ago. Note how she is about one-third of the way into the photo, and the waterline is roughly one-third of the way down from the top. I’ve also included a smaller image that shows these one-third lines. Imagine how poor image would look if Tanya’s face was dead center in the frame. There would be a ton of boring, featureless sky above her head, and she would be a lot smaller to keep her whole body in the picture.
Another thing to note is that you don’t have to hit the one-third lines and intersection points exactly to have an effective composition. In the example with Tanya, the waterline is a bit above the one-third line, but it still works well.
That’s basically it. Avoid bulls eyes and use the rule of thirds, and your photos will look a lot better than the average snapshot.