I was reading an article recently on Popular Science about an algorithm that can stitch together mobile phone video footage of the same event into a single, synchronized, multi-angle film. Here’s the description from the article:
"As we upload more and more videos to the Internet—one hour of new video every second to YouTube alone—experts are finding new ways to mine them. A team led by Igor Curcio of Nokia’s Research Center, for example, has developed an algorithm that stitches concertgoers’ cellphone footage into a single, synchronized multi-angle film. The concept is relatively simple: the audio track serves as a guide to sync up the footage, and the software chooses the best shots. Curcio has no real business model yet—photography is prohibited at most concerts—but giving people the ability to identify and coherently connect common elements in multiple videos is nonetheless a step toward something significant."
As I read this article, I thought it was pretty cool. I don’t imagine it’s a very complicated algorithm, assuming that you know the videos are all from the same location (many photos and videos have timestamp and even potentially longitude and latitude information stored with the file), aligning the video streams along the sound track then selecting (whether manually or automatically) the ‘best’ image/angle to use for a particular time slice shouldn’t be that hard. I imagine that with this technology we’ll start to see some VERY interesting videos produced from all sorts of crowd events.
I saw an automobile commercial a while back where the manufacturer had hundreds of folks line up along a car’s route and film the car with their smartphone cameras as it goes by – then stitched the video segments into a short video. Movement along a path with multiple angles I imagine looks pretty cool.
Imagine someone arranging this for a concert or even a theatrical presentation – if you can get 10, 20 or 100 people to hold their smartphones up for the entire performance or scene you’ll have a modern, cool format to present the show to others. You could use videos from folks further away from the stage for transitionary segments or slower moving parts of the production or song then ‘zoom in’ to closer cameras for the important stuff. I’d love to see this done more – I bet some innovative folks can do some pretty interesting stuff with this technology.
Now, I’m not a paranoid person (although just because you’re paranoid doesn’t mean people aren’t out to get you), but the next important part of the article really scared me:
"For instance, the drones that patrol the U.S.-Mexican border and the security cameras in cities already record more footage than human observers can possibly examine. If an agency could rely on a computer to track individuals, groups and events on its own, agents could use intelligence far more—well, intelligibly."
We’re hearing more and more about law enforcement organizations using social media such as Facebook and Twitter to identify criminals from event images and social media posts. That the technology described in this article could be used for tracking people through a city really scares me. When you think about it though, they don’t need any special algorithms to do this today. Many governments already have the facial recognition software they need and multiple camera angles to pull from to create sophisticated videos from crime scenes. Cameras are everywhere, put there using the paranoia generated by the September 11, 2001 terrorist attacks in the US. Assuming someone could coordinate the timestamp of multiple videos (clocks on video recorders are set manually today right? Enabling them to be set via atomic clock would make this easier), we’re not far away from being able to see everyone in a crowd from multiple angles after a terrorist event or serious crime.
The good news is that this technology will make it easier to catch criminals and terrorists. The bad news is that these cameras are everywhere and there is very little to keep governments (or even criminals) from watching your every move. For any Rush fans out there – this is exactly what’s described in Red Barchetta.