headshot: my first production photo project

The following post is a compilation of notes from my first production photography project. The goal was to take a headshot, appropriate for a reporter, but without access to either a studio or a light kit.

Not having access to a studio and never doing a headshot before, I turned to the web to do some research on what I could do. Photo.net had a good article on portrait photography, particularly the section on Stealing a Location.

What if you don't have a big open space with diffuse light and a neutral background? Steal one. If you live in the United States, a vast open space with light pouring in from expensive skylights is as close as your nearest art museum or university. With a 200mm lens set to f/2.8, the background will be thrown out of focus.

Unfortunately, most of the museums in DC do not have windows at all -- neo-Classical architecture is all about columns. There is one Modernist museum in DC though, I.M. Pei's East Building of the National Gallery of Art. The East Building has an atrium with pink marble walls (really a light gray limestone with a warm hue) and is topped with pyramidal glass skylights, which are screened by aluminum rods to reduce the sunlight. The photo below is a picture of the wall with light projected from the skylight, with shadows from the screen of aluminum rods appearing as fine vertical lines.



Cool light pattern on the walls -- I'm guessing that was an intentional effect by the designer. My intent however was to blur the walls by setting the camera to a wide aperture (smaller f-number) so the patterns in the stone, the joints, and the light patterns become non-discernible and non-distracting but provide some variation in the background so that it's not just flat gray. Something like in the following photo.



The shoot was scheduled for a sunny afternoon to get the most light coming through the skylights. I arrived early to take some test shots on my own to check the light levels at specific locations in the atrium. I found the best lighting was not on the ground floor lobby, but rather on the highest bridge across the atrium. The bridges blocked some of the light from reaching the ground floor, so it appeared slightly darker compared to being on the bridge where it was as close to the skylights as we could get. Plus, the highest bridge in the atrium led to an empty/closed exhibit space so there was no foot traffic at all on it.

I put so much energy into light, background, and color that the camera settings seemed relatively simple. I used a 50-150mm f/2.8 telephoto zoom lens on my Canon Digital Rebel XT (equivalent to a 70-200mm on a full frame camera). Using the longer focal length of a telephoto lens results in flatter perspective, which generally results in more flattering portraits and better background blurring. I set the camera to ISO 200, evaluative metering, aperture priority mode, and set the aperture at f/3.5. The shutter speeds the camera was choosing (around 1/500 sec) seemed to make sense and the test shots looked fine. Lastly, I put the camera in continuous shooting mode so the camera would keep snapping for as long as I held the shutter release button down. I like snapping three or so photos at a time to account for changing facial expressions.

After meeting up with J, we started with a session of about 50 photos. I was fortunate that she handled all the things I knew nothing about: hair, wardrobe, and makeup. In addition, she had a good vision of what she wanted. I had a good general idea of what a headshot for a reporter should look like, but I didn't know how to direct the poses to get to that point, so one could say J self-directed her poses. After the 50 photos, I fired up the laptop to start looking at what we had. (The camera's LCD screen is just way too small to tell if the focus and details are correct.) After narrowing the choices down and seeing any quick improvements we could make, we went back out on to the atrium bridge and shot about another 50 frames. Back at the laptop, we found one that stood out from the rest and it ended up being our final selection. All in all, it took about an hour -- much faster than I originally thought it would take.

Later back at home, I tried a few crops and decided on a square crop. I experimented with adjusting some of the levels and the contrast/brightness, but I found that I was happy with the original exposure and that the image looked best without any additional post-processing. The below photo is the final result.


83mm focal length (133mm equivalent on a full frame)
f/3.5, 1/400 sec, ISO 200, handheld, no flash

Lessons I learned from this project: (1) light and background are the dominating issues when taking portraits outside of a studio, (2) I need to pay more attention to poses and expressions to eventually get better at directing, and (3) bringing a laptop to location is a great way to check results and make sure everyone is happy before leaving.

In a broader sense, I'm realizing that photography is very much about proper planning and attention to detail. The smallest shadows make a huge difference in the end result, but one has to be able to recognize issues and then plan and/or react accordingly to resolve them. Strangely enough, I'm carrying lessons from software engineering (my day job) into photography. Sort of makes sense as the smallest of software bugs can create the largest problems. Pay attention to the user/customer's needs, address the greatest risks first by experimenting/prototyping to understand them well enough to mitigate them, plan on at least two iterations to end up at the final product, while re-evaluating after each iteration. Thankfully, digital photography lends itself well for quick feedback, making iteration feasible :)