Skip to Content

Algorithmic Video Editor Turns Amateur Athletes into GoPro Heroes

Automated video and sound editing software can make any bumbling idiot into an extreme-sports movie star.
August 21, 2015

Visit any surf break, mountain bike trail, or ski resort and you’ll see people diligently filming their best moves and worst falls–and a lot of tedium in between–using smartphones and action cameras such as GoPros. Startup Shred Video has developed technology that tries to edit collections of such footage into short, slick, and shareable movies at the push of a button.

The company’s software uses algorithms that analyze and edit or remix video and music to try and make short movies in the style of those produced by extreme-sports brands to show off the exploits of their sponsored athletes. The final results are generally two minutes long or less. The company’s software is available to download for Apple computers.

Shred CEO Mike Allen says the company wants to be something like the video equivalent of Instagram, which by providing photo filters made it easy for anyone to make mobile snaps more interesting to other people. However, because producing video is more complex, Shred has had to develop more sophisticated technology. “It takes more than a sepia filter to make video great, it takes curation, pacing, synchronized audio and video,” said Allen, launching his company at an event held by the YCombinator accelerator program on Wednesday.

Software isn’t capable of creative decisions, but Allen says the relatively formulaic structure of extreme-sports videos is within its grasp. Shred’s software sets out to make movies that begin with slow-moving establishing shots (for example, showing the beach or ski slope at the start of a trip), transition to highlights of the adrenaline-pumping action, and then wind down with more shots of scenery and people at the end. “That formula still works even though you’re not doing the most extreme back flips,” says Allen.

The software selects footage to use by looking at time stamps and the pattern of acceleration in the frame. It can also shorten and remix any song provided by the user, using algorithms that can identify and edit the different musical parts of a track. The software aims to play the intro to a song during the establishing footage, and then switch to the chorus or a more up-tempo section just as the action begins. Cuts between clips are all made on the beat.

Last year, users of Facebook and Google’s social network, Google+, were offered automatically generated “year in review” movies combining both photos and video. They used music but didn’t let you choose it, and didn’t appear to remix it to fit the visuals. Google’s Photos service automatically creates movies for, say, vacations and allows you to choose from a menu of music and some visual filters, but it is not focused on action footage.

Allen concedes that his more ambitious software doesn’t always get it right the first time, but he says that it makes it easy enough to provide quick feedback—for example, to include or leave out some footage. A slick 2.5-minute video Allen made of a surf trip with friends took less than five minutes of work to produce from a collection of more than six gigabytes of raw footage, he says.

Shred will currently edit your video for free but charges $1 or more to save it.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.