Tired of watching hours of B-roll footage to find your shots? With a new highlight-identifying technology, Pik'd is looking to change the way you view your GoPro footage.
pik’d – the brain for your GoPro. from pik’d on Vimeo.
Perhaps because my own personal GoPro footage consists largely of bails, flails and otherwise unshotworthy events, I’m well aware of the difficult task many GoPro users have in sorting through minutes and minutes of uninteresting footage in search of the best action. For those who can’t stand another hour of sorting through random clips, Pik’d is a new hardware/software tool aimed at helping you get the most out of your GoPro footage, quickly.
Like some other third-party products on the market, pik’d can help you access and post your GoPro footage on the go via mobile app. But the product’s main feature is a proprietary technology for identifying highlight moments in video clips and sorting them out from the rest. That’s right — pik’d claims to be able to identify that awesome trick in an otherwise lackluster run, and make it quickly available for sharing or exporting.
Here’s how it works: the pik’d hardware element, a low-profile motion tracking device, attaches to the back of your GoPro, and collects a range of precise motion data whenever the camera’s on. Then, when you connect your GoPro to your computer (a mobile app is also in development), pik’d software syncs your video with the motion data, and uses algorithms to analyze the data and determine what’s a highlight in your video, and what’s not. The “highlights” as chosen by pik’d are clipped and made available for exporting or sharing, without the time-intensive work of sorting through all the rest of the footage.
“It’s literally a 10 seconds – 3 clicks process from connecting your GoPro to sharing your footage,” says Pik’d co-CEO Clemens Haaser.
While the fast turnaround for social sharing is certainly a highlight, the product’s success hinges on the dependability of it’s highlight-identifying algorithms. The company collected thousands of hours of motion data from several action sports, and used machine learning techniques to determine what kind of data likely corresponds with a “highlight” event. The resulting algorithms, pik’d says, are able to dinstinguis between when you’re casually skiing down the bunny slope, and when you’re mid-way through a double-cork 12.
“For surfing we have reached an accuracy of 95%, so no wave or wipeout remains unseen,” says Haaser. “We are on the way to reach the same level for the other entry sports: Skiing, Snowboarding, Downhill and Motocross, so our users can totally rely on pik’d.”
Pik’d will hit markets soon with a retail price of $99USD, but the company’s offering a limited-time sign-up via their website to buy pik’d for $60.
So if your GoPro’s always on, but you don’t have the time to edit your footage—or if your banger-shot to boring-moment ratio is as bad as mine—then you might consider pik’d a way to optimize your GoPro experience. Because none of us needs to endure watching all the way through an extremely boring run filmed from top to bottom again.