How do you make an interesting and dynamic photo when all you have to work with is ugly fluorescent light?
For the last few years, I’ve been photographing cosplayers at the Montreal ComicCon. The costumes are often inventive and visually arresting. But from a photographer’s point of view, the lighting conditions at your typical convention center are appalling.
Usually, convention halls are lit with inexpensive fluorescent tubes, which cast a distinctly greenish glow. The light is also top-down, putting shadows on people’s faces in unappealing ways. It’s an ugly light, and it works against you as you try to make a dynamic photo.
Now, it’s certainly possible to correct for the greenish hue of the fluorescent light, and return skin tones to normal — either using image processing software, or simply by selecting a white balance setting on your camera that matches the light in the room. But even then, the light is still unflattering and boring.
Check out the photo on the right.
This photo is… okay. At least it is correctly exposed. But note the unpleasant shadows in the man’s eyes and the rather shapeless quality of the light. Most importantly, the subjects’ faces don’t stand out from the background very much; everything on the convention floor is lit evenly. It looks like it was shot in someone’s office.
So this year, I wanted to experiment with a different technique, and make photos with a bit more zip.
Flashing the Subject
The only way I could ameliorate the unpleasant light in the convention hall was to add more light into the photo, light that I could control myself. This meant using a flash. But putting a flash on top of the camera would produce a paparazzi effect, which would be worse than using no flash at all.
The photo at right shows the worst-case scenario for on-camera flash. Also note the incredible red-eye effect.
To add more light into the scene without making the photo worse, the light source must be located some distance from the camera, and it ought to be diffused and softened in some way.
I toyed with a few options: an on-camera flash with a small diffuser, or a small softbox on a moveable stand. But it came down to my need for small size and portability. Conventions are crowded, with people moving around all the time. I needed to move quickly and fluidly, and keep out of peoples’ way.
I settled on the setup you see in the picture on the right.
The white plastic thingie on top of the flash is a Gary Fong Lightsphere diffuser. It softens the harsh light of the flash, spreading it more evenly, and creating more pleasing shadows. It’s not as good as a softbox for making pretty light, but it had two advantages: first, it was small and lightweight; and second, I didn’t have to worry about which direction it was pointing.
So when I started shooting, I worked two-handed: the camera in one hand, and this Gandalf-esque magic wand in the other. I raised or lowered the flash as I judged was appropriate for the subject.
Exposing the Subject
Now, I didn’t want the flash to overpower the ambient light in the room; that would create a stark image, with a harshly-lit subject against a dark background. Rather, I wanted to mix the ambient room light together with the light from the flash. This, I hoped, would provide some direction for the light.
So, I set the camera to underexpose the scene by about one and a half stops, which would make the whole picture darker. The subject would be illuminated by a mixture of light from the flash and the overhead fluorescents. This would make the subject brighter than the background, creating a more dynamic and interesting picture. At the same time, by holding the flash out to one side, the light would be more flattering and make the subject come alive.
This is the result:
I think this is a marked improvement.
I decided not to use my SLR for this shoot. Wielding a camera that weighs 1.5 kilos using only one hand for several hours was not going to be comfortable. Instead, I opted for my Fuji X100, which is small and relatively light, but has a terrific sensor and lens. Critically, it also features a hotshoe so I could use it with a radio trigger for the flash.
The X100 has a fixed lens — no zoom — with a field of view suitable for street photography (focal length 23mm on a APS-C sensor). This makes it ideal for getting head-to-foot shots of the cosplayers, but not so great for closeups. Getting close to the subject with a wide lens produces unpleasant facial distortions. Still, the lack of zoom makes it easier to operate one-handed.
For some shots, I held the light out to one side as far as I could, For other shots, I felt that an overhead light provided drama. My choice was dictated by the costume and character.
Lessons Learned
I made one serious mistake with these photos, which required more post-production work than I would have preferred. Remember when I said that fluorescent lights have a green hue? Well, what I should have done was to buy a green coloured gel to place over my flash so that the subject and the background would have the same white balance. I forgot to buy the green gel, and said to myself, “Well, how bad could it be?” As it turned out, pretty bad.
The image at right shows the greenish hue of the fluorescents without correction in Lightroom.
So, for most of my pics, I had to correct the colour of all the backgrounds so they would have the same white balance as the subject. Having done all that work creating masks to separate out the subjects, I also took the liberty of reducing the saturation of the backgrounds by about 40% in order to make the colours of the costumes pop a bit more.
Next year, I’ll be sure to buy a variety of green gels in advance.
I may also try shooting with two cameras, both with fixed lenses. As I mentioned, the 23mm lens on the Fuji X100 makes getting flattering closeups difficult. A second camera with a 50mm or 85mm lens might allow me to switch up and get a wider variety of shots. Maybe I’ll experiment with a softbox next time, too.
Or maybe I can bribe someone into being my mobile light stand. Any takers?
If you curious, have a look at my whole album of pics from the 2014 Montreal ComicCon on Flickr.
How did you decide to frame them? Pose them? Thx
ShooterMike: When shooting the cosplayers, I usually had about 30 seconds to take a photo. The better-costumed attendees are always swarmed with people taking pictures, so I wanted to limit my own imposition on the cosplayers’ time. Also, I always ask permission before shooting.
So, framing is done on the fly and very quickly. I always tried to clear the background as much as possible, to isolate the subject. I may ask the subject to take a few steps to one side, or tell them that I’m going to wait five seconds while someone behind them clears the shot. Given that I was shooting with the X100 and its wide field of view, head-to-toe shots were easy, and a natural choice given that most costumes are worth seeing in their entirety.
I took two or three frames of each subject, thanked them, and allowed them to move on.
As for posing, most cosplayers have developed their own character poses. Most of them already know what position suits their look. This is a real blessing, as it allows the shooter to work more quickly. I may ask the cosplayer to raise or lower a prop to keep it in frame, or tilt their head for a better expression, but little else.
Some photogs will seek out specific cosplayers, the ones with the most spectacular costumes, and ask for a private session in a quieter corner of the convention center, or outdoors. This way, you get to control the lighting and background, and the results are much better. You get to shoot fewer subjects this way, and it requires a rigorously professional attitude in order to make sure the subject is comfortable. And if you bring lighting stands and umbrellas, it means more equipment to lug around.
Kay. Thnx.