When violence breaks out at a protest, a system that automatically synchronises video of an event from different cameras can give prosecutors clear evidence
MILLIONS of people took to the streets in Brazil last week across over 100 cities to protest against wide-ranging issues, including corruption and poor public services. Like the ongoing unrest in Turkey, such demonstrations can provoke violent crackdowns from security forces. More than ever before, the turmoil is being caught on camera, by news agencies and demonstrators alike. Shot amidst chaos, these videos and images are often uploaded devoid of context, commentary or vetting.
The trouble is a single video merely gives viewers a snapshot of an event, says Christoph Koettl of Amnesty International, and snapshots can be misleading. New tools are changing this, putting images and video of dissent in context. They have the potential to catch those who violate human rights in the act, making it possible to prosecute the perpetrators.
One such tool is called Rashomon, built by Ken Goldberg and researchers at the University of California, Berkeley. The software automatically syncs video from multiple sources shot from different angles, using metadata on the files – the date, time, type of equipment used, and information about the exposure – if that is available, or if not, visual and audio cues. Failing that, humans can also manually sync the videos.
One chilling compilation charts protests on 31 May in Istanbul, Turkey. It shows a man screaming at an armoured van with a water cannon mounted on top. He charges at the vehicle and kicks it, then moves back into the street, still yelling. Without warning, the water cannon opens up, knocking him off his feet. He lands sickeningly on his head and doesn't get up. From another angle people are seen rushing to the man's aid, only to get sprayed themselves. Another angle clearly shows an identification number on the van.
"With Rashomon you might actually see who is shooting from a specific direction," Koettl says. "Who was responsible? Who should be held accountable? This comprehensive view could be very, very helpful in our investigations."
The system's main limitation is that YouTube strips metadata out of videos, says Goldberg, so it must rely on audio cues to sync videos – but this only works when there is a loud noise easily distinguishable from the background, like a gunshot. Koettl says the metadata issue may be resolved by using a new app called International Evidence Locker. It lets a witness take pictures and videos, then encrypts them and sends them to a secure server with all the information intact.
Bonnie Freudinger, an engineer at the Medical College of Wisconsin, who works on the app, says its purpose is to provide a clear chain of digital evidence, which makes it much more valuable in court proceedings. "The data will be sent to two locations," Freudinger explains. "You could send one to a server at Amnesty, and one to the International Criminal Court. They can be decrypted and compared at trial for verification."
Koettl foresees Amnesty using both tools in conjunction to compile evidence of human rights abuses. "The way we approach human rights investigations is changing," he says. "You want the most complete view possible – different angles and different viewpoints." We will never get a full view, but with a tool like Rashomon we are on the way, he adds.
This article appeared in print under the headline "The straight story"
- Subscribe to New Scientist and you'll get:
- New Scientist magazine delivered every week
- Unlimited access to all New Scientist online content -
a benefit only available to subscribers - Great savings from the normal price
- Subscribe now!
If you would like to reuse any content from New Scientist, either in print or online, please contact the syndication department first for permission. New Scientist does not own rights to photos, but there are a variety of licensing options available for use of articles and graphics we own the copyright to.
Have your say
Only subscribers may leave comments on this article. Please log in.
Only personal subscribers may leave comments on this article
All comments should respect the New Scientist House Rules. If you think a particular comment breaks these rules then please use the "Report" link in that comment to report it to us.
If you are having a technical problem posting a comment, please contact technical support.
