Preventing Movie Piracy
Researchers are developing tools to thwart the copying of films in theaters.
Last month, we ran a story about an experimental device for locating and blocking cameras (“Lights, Camera – Jamming”). Here we look at another set of technologies aimed at dissuading would-be bootleggers.
The movie industry has a problem. According to the Motion Picture Association of America (MPAA), Hollywood loses billions of dollars a year on illegally sold copies of movies. The losses are impossible to calculate accurately, of course, since it’s unclear how many of the people who download copies over the Internet or pay a few dollars for a pirated DVD would have paid $9 to $20 for a theatre ticket or legal DVD. Nonetheless, studios and theatre owners are eager to find ways of safeguarding their intellectual property.
Paris-based Thomson, which provides technology to the entertainment and media industry, is exploring methods for thwarting at least one type of bootlegger: the covert camcorder user. The company’s technique involves inserting “artifacts” – extra frames, flashes of light, or pixelated grid patterns – into a movie during its digital-processing phase, before it’s shipped to theatres. The goal is to mar a camcorder recording without degrading the images moviegoers see, says Jian Zhao, chief technology officer of subsidiary Thomson Content Security in Burbank, CA.
The artifacts exploit the differences in the way a human brain and a camcorder receive images. In the technique that’s furthest along, extra frames – with the words “illegal copy,” for instance – are inserted into the film. These warning words flicker by at a frequency too fast for the human brain to process – yet they appear in a camcorder recording.
This difference is possible because movies are projected as a series of still shots. Film projectors flash 48 images per second (24 frames are collected each second, but each frame is flashed twice) and high-end digital projectors can flash even more, according to Thomson researchers. The limit for human visual processing is around 45 flashes per second; above that, a flickering image appears continuous. Furthermore, camcorders do not average frames, as eyes and brains do. Instead, they’re sampling devices that take a series of snapshots – collecting many more frames per second than our visual systems. Hence, frames that eyes would miss show up in a camcorder recording – and are reproduced on a video screen when the recording is played.
Using extra frames to obscure a recording isn’t as straightforward as it seems, however, since camcorders could theoretically be set to a sampling frequency low enough that they’d miss the hidden message, says Zhao. That possibility requires counter-countermeasures, such as randomly adjusting the frequency at which the extra frames appear. Camcorders can’t yet adjust their sampling frequencies quickly enough to keep up and produce a quality recording. But camcorder technology will continue to evolve, says Zhao, “and thus, we’ve got to evolve.”
In addition to the frame-insertion technique, Thomson is working on incorporating additional sabotaging mechanisms into its system, such as projecting ultraviolet or infrared light onto the screen and washing out camcorder pictures. Aware that the easy counter-measure to this is simply to place a filter over a camcorder’s lens, Zhao says their system is being designed to combine many different wavelengths, so that finding the perfect filter would be difficult.
Additionally, the researchers want to take advantage of interference patterns that can be created by overlaying film frames with grids of tiny features that are too small to be resolved by human eyes. Overlaying two of these grids at a certain angle creates a moiré pattern that a camcorder picks up, but people miss.
An anti-piracy system that included some or all of these techniques could make it prohibitively expensive for bootleggers to keep pace.
By altering the movie itself, says Zhao, their system avoids some of the potential drawbacks of other anti-piracy prototypes. Some systems, for example, position cameras in front of an audience, where they actively hunt for the distinctive reflections given off by CCDs, the light-sensitive chips that capture images in a camcorder, and send beams of light that temporarily disable the chips (see “Lights, Camera – Jamming”). Thomson’s system would be less intrusive, says Zhao. “I would not feel comfortable with a camcorder constantly monitoring the audience,” he offers.
Even with the progress in such technologies, however, it could still be years before an anti-piracy system becomes commercially viable, says Zhao. “A lot of people in the studios have some doubt if we can ever have an effective solution,” he says. In March, Thomson opened the Burbank Innovation Center to keep Hollywood organizations up to date on its progress and to garner feedback, according to Zhao.
Ultimately, acceptance of such a system will depend on several factors, says Ethan Bush, senior project director at National TeleConsultants, an engineering and design consultancy for the media industry. For one, on-film artifacts must be completely hidden from audiences. For another, the rapid play of words, light, or patterns across a screen cannot have harmful side effects. “We don’t want anyone going into epileptic seizures,” says Bush. Yet, he adds, piracy is “a huge issue” and an effective solution could be worth billions.