Computer Generated Imagery Experts Say Man Flying Like a Bird Video Is a Fake

Was Jarno Smeets' video showing him flying with these man-made wings a hoax? (Photo: Human Birdwings)

Yesterday, we brought you the video of a dream turned reality for one man who built his very own set of wings and soared in the sky like a bird. Now, computer graphics experts and others looking into Jarno Smeets’ resume are saying it could be a hoax. If it is a fake though, they’re all pretty excited about how well-done it is.

(Updated: Man who flew like a bird in viral video admits farce was part of an online experiment)

Smeets, supposedly a mechanical engineer from the Netherlands, began extensivly chronicling the construction of his DIY wings in August 2011 on his blog Human Birdwings. Since then, Smeets was featured on sites like Wired, TechCrunch, the Discovery Channel, and the Financial Times in Germany, among others. While news about his flight efforts began picking up interest on the Web late last year, the success of his second test flight helped his story take off this week.

Computer Generated Imagery Experts Say Man Flying Like a Bird Video Is a Fake

Jarno Smeets -- or is it? (Photo: Human Birdwings)

Now, Wired reports, some facts are not checking out and elements of the video seem fishy. Wired hosts a slew of computer-generated imagery experts who share why they think the video might be a fake:

Ron Fedkiw, a computer scientist at Stanford University who has worked on computer-generated graphics in films such as “Terminator 3″ and “Star Wars: Episode III,” told Wired in an email that a continuous video shot might have made Smeets’ latest video “a much more convincing fake.”

“[C]utting the camera angle is an obvious trick,” Fedkiw wrote to Wired in an email. “Note how there is no continuous video from take-off to landing, instead they cut away the main ground camera right as he takes off and cut back right before he lands.”

“They don’t really even need much CGI work or any fake footage here with the camera cuts,” he wrote. “The head cam footage could all be shot from a glider video — any glider, not necessarily those wings … The only real image work would have to be in the very beginning when they get a small bit off the ground, which could just be running up an edited hill or ramp.”

Watch the footage of the second test flight, which has earned more than 1 million hits on YouTube since its March 19 posting:

Gizmodo brings in Ryan Martin, technical director at Industrial Light & Magic, and his colleagues who are currently working on the film “The Avengers.” Here are some of their thoughts (Note: Check out Gizmodo’s full post to see all 11 of experts quoted):

Okay, so I don’t see any glaring visual problems, but that’s expected when the quality is as sh**** as this. But that’s the first thing that makes me question its authenticity. They’re able to afford to build this thing, but can’t invest in proper video equipment, or… a tripod. If I were to make a fake video with the intention of going viral, I would make certain that the quality was as poor as possible to disguise any flaws in poor cg work. Another big visual issue I have with this video is the stability of his head during flight. Try and keep your head that still while waving your arms up and down when they aren’t attached to a giant wing contraption. Still, it seems almost too crazy to be fake and I was unable to find other glaring flaws with the video. So, I’ve queried our entire facility because we have some pretty amazingly smart people here. Here’s what some of them had to say:

Employee 1: ”without a doubt, fake”

Employee 2 (also a pilot) “the camera seems very strange. I know that when I am flying in an airplane, I don’t look straight ahead all of the time. Also, the only way people have been able to propel themselves above the ground have been by bicycle arrangements to power a fixed-wing aircraft. A human powered helicopter managed 10 seconds of flight about 5 inches above the ground. The legs are much more powerful than the arms.

I have serious doubts about it just on the physics and physiology points alone. ”

Employee 3: ”I agree, I saw that earlier today. I can’t spot any glaring visual problems, but the physics just don’t add up.”

Employee 4: ”Bad physics, shaky cam with bad focus (always a giveaway) and the most steady head I’ve ever seen on a guy flapping his arms in order to not break every bone in his body. FAKE.”

Computer Generated Imagery Experts Say Man Flying Like a Bird Video Is a Fake

Inconsistency in the fabric seen around 1:45? (Image: Gizmodo)

What’s more, Gizmodo reports Martin saying he thinks he has proof it’s a fake. When you watch the video, you can see a little black square on the left side of the fabric at 1:45. Here’s the kicker:

Now, without cutting, the camera pans down and then back up again. When the camera pans up, the wing is cg. You can tell because the model they used didn’t have perfect textures

Martin also found an issue with the shadows at 0:18:

At 0:18 — the cast shadows of the three fellas on the wings are another giveaway. The shadow on screen right, for example, was created using an articulate of the man himself, then hand-animated and warped to look like a cast shadow on the wing. Watch how the cast shadow does a moon-walk/shuffle, incongruous with the man himself.

All that said, it’s still very well done. Good stuff. And I truly believe this is a healthy exercise for folks who do this stuff for a living.

Gizmodo has another expert who is convinced it’s a fake but shares excitement over how well-done the video is:

This is 100% without a doubt a digital composite, which is great! When I saw this video I was happy to see that somebody really took the time to integrate good CG into a viral video. So many hackneyed attempts are passed around, and this one really stands out.

Still a hopefull believer or have your flight dreams come crashing down? If you’re still hanging on, Wired also reported some issues confirming facts about Smeets himself. Wired contacted Coventry University, where Smeets supposedly attended from 2001 to 2005, and found they had no record of his attendance. It also contacted some employers listed on Smeets’ LinkedIn profile: one responded in time for publication stating they also had no record of an employee by that name.

Smeets also said in one of his blog posts that he ran his ideas by neuromechanics scientist Bert Otten at University of Groningen, which Wired confirmed with Otten did in fact happen. But here’s what Otten had to say:

“I haven’t seen the contraption they have built with my own eyes, so I cannot tell you from the inside whether this is fake or not,” Otten told Wired. “I haven’t looked at the video very carefully, but others have, and I must share their suspicions there.”

Wired states that it contacted Smeets who responded via email but denied them a phone interview because he was already bogged down with media requests.