Skip navigation

Tag Archives: video

good article on the now-famous esquire cover shoot..

Frame grabbing: The art of drawing great photography from video

By Edward J. DelaneyMay 810:21 a.m.

The June issue of Esquire arrives on newsstands Sunday, and there’s something unique about its cover photo. Not the presence of an attractive young starlet — that’s de rigueur in the magazine business. It’s that the photo of Megan Fox was shot with a video camera, not a still one. Photographer Greg Williams shot footage of Fox with one of those jaw-dropping Red One cameras and pulled stills from the resulting video. (As Zach noted recently, Esquire seems to be cornering the market on cover gimmicks: e-ink, mix-and-match flip books, and now framegrabbing.)

I thought that was a perfect reason to post this interview Ted did with Pulitzer-winning photographer (and my former coworker at The Dallas Morning News) David Leeson about frame grabbing — an area where he was an early innovator. —Josh]

To David Leeson, the appeal of frame grabbing seems obvious. It reduces the number of tools a photographer has to juggle, and it enables multiple outputs from the same journalistic workflow. As he wrote about his first experience, preparing to cover Hurricane Katrina in 2005:

The first thing I did upon receiving an HDV camera…was shoot a few seconds of video, import it with iMovie and make a frame grab. The results were almost as magical as the first time I saw a print emerge in a tray of developer. I knew the world of photojournalism as we knew it, would never be quite the same again.

But he’s been surprised by the resistance among many of his fellow photojournalists. Even as each wave of new cameras to hit the market makes frame grabbing an easier option, Leeson still finds himself preaching to the unconverted. The main resistance may be the core belief that the fundamental art of the photograph is timing the decision of when to press the shutter.

“We all know Cartier-Bresson’s famous decisive moment,” Leeson says. (Here’s Cartier-Bresson in 1957: “There is a creative fraction of a second when you are taking a picture. Your eye must see a composition or an expression that life itself offers you, and you must know with intuition when to click the camera. That is the moment the photographer is creative.”)

“But we also fail to realize that the decisive moment wasn’t possible without the technology -– the 35-millimeter format, and Leicas and Contaxes that were actually technological developments that were pooh-poohed and laughed at in the age of 4×5 Speed Graphics.” Despite the technology, you’re still a photographer, he says: “Nothing has changed but the camera in your hand.”

To newsrooms where frame grabbing has meant occasionally taking a grainy still off recorded television, the term may not evoke high quality. But in the last five years, digital high-definition camcorders have advanced to the point at which they can capture, at 24 or 30 frames per second, full-quality images. The Red One used in the Esquire cover shoot shoots the equivalent of a 9.4-megapixel still 24 times a second. In one sense, that may not seem too different from motor drives on high-end still cameras that can shoot at high speeds, like the 11-frame-per-second drive on the Nikon D3. But it’s oceans away philosophically.

The push toward frame grabbing has several origins:

— More and more news photographers are being asked to bring back video and stills. With frame grabbing, that can involve one shooting process, not two.

— New “still” cameras — like the Nikon D90, the Canon EOS 5D Mark II, and the upcoming Panasonic Lumix GH1 — have remarkable video capabilities, and look and feel like the equipment photojournalists are accustomed to working with.

— Many new HD camcorders (such as the new Sony PMW-EX1) are getting smaller. They’re also meant to be shot at the eye (like still cameras) or at the chest through an LCD viewer — not on the shoulder. Both still cameras and camcorders are coming out with bigger, sharper, brighter LCD viewers that may eventually eliminate the need for the traditional eyepiece.

— Some camcorders just over the horizon which promise resolution at significantly higher rates than current HD, at a lower cost and in a smaller package. The Red One, which starts around $17,000, shoots video four times sharper than current HD, and Red is promising smaller, significantly-less-expensive models in the future.

But Leeson warns that shooting video involves a different set of imperatives than shooting video for the purpose of grabbing frames. He says he can tell the difference between video shot for stills and video shot for video. To see an example of what he’s talking about, check out the documentary film War Photographer, in which filmmaker Christian Frei attached a small videocam to photographer James Nachtwey’s still camera. Since Nachtwey was shooting film, the video gives a sense of how this difference plays out. Leeson notes that video seeks “a beginning, middle and end,” while shooting stills privileges “a moment.”

For photographers looking to use frame grabs from video, here are some considerations:

Frame grabbing works better on a camcorder with a “progressive” shooting option. In progressive shooting, whole frames are shot one at a time — just as in traditional film shooting. The old television and video standard is “interlaced”, in which a new frame begins to scan from the top of the frame as the previous frame moves toward the bottom. Until a few years ago, camcorders shooting progressive frames were rare and expensive. Now, many camcorders can shoot either 24p (24 progressive frames per second) or 30p.

Frame grabbing works better at higher shutter speeds than would be used for shooting video. A rule of thumb for shooting 24p video is to set the shutter speed at 1/48, which is much slower than one would use for normal still photography and yields somewhat motion-blurred frames that work well for film. For a still photographer at a football game, for example, shooting at 1/1000, 1/4000 or even higher is common; shooting at those speeds in video gives a jerky sensation to motion. It’s worth noting that the EOS 5D Mark II sets video shutter speeds between 1/30 and 1/125, which will make it less capable as a frame grabbing option for sports or other fast-moving subjects.

Frame grabbing might require a video light rather than a strobe in low light. Small camera-mounted video lights give decent light, but usually fairly directional. The new combo cameras such as the EOS 5D Mark II don’t apparently combine strobe with any continuous light for video.

Gain settings substitute for film speed settings when shooting video. Understanding how those relate takes some experimentation.

Processing frame grabs involves several steps. Typically, the video is opened in Final Cut or iMovie and stretched to its largest on-screen size. (The bigger your screen, the better the final image.) A screenshot of that image is taken, and then the shot is converted to a JPEG in Photoshop or a similar image-editing program. Leeson’s son, David Leeson II, also developed a tool called Voodoo that helps process the screenshot image using algorithms to accentuate the image before the compression to JPEG. (The Digital Journalist’s James Colburn critiqued frame grabbing here).

Frame grabbing is also possible from such simple devices as the ubiquitous Flip camera and its peers like the Aiptek. As small videocams get better and better, frame grabs will become an increasingly viable option. Meanwhile, Leeson — who accepted a Dallas Morning News buyout offer in September — has migrated increasingly into documentary filmmaking. When I spoke to him, he had just finished the edit of At War, which he produced and which is directed by photojournalists-turned-frame-grabber-turned-filmmaker Scott Kesterson — perhaps suggesting that frame grabbing will serve as a bridge technology for still photographers moving into HD film and multimedia projects.