Intelligent Surf Cameras

Ben Freeston
Surfline Labs
Published in
3 min readJan 23, 2019

Surf cams changed the way we plan (or don’t plan) a session. Your eyes on every spot, in real-time, all the time. But your camera relationship is a demanding one. You have to pay constant attention if you don’t want to miss something and cameras can tell little lies; you’ve viewed five minutes of video but still missed the set. Add in checking a handful of spots and you could be spending a healthy amount of time staring at the surf on a computer screen, rather than actually surfing.

The answer is an intelligent surf camera. One that knows what it’s looking at and can turn that video stream into some hard data for you. The sort of data that can create the sort of alert that means rather than you paying attention to the camera the camera is paying attention for you.

Counting surfers and detecting surfing with the Surfline El Porto camera

We approached this project wanting to start with the single most meaningful event to a surfer. Surfing itself. Sure you want to know the size of the waves, the tides and the wind, but if you knew that right now a handful of people were scoring at your local spot odds are you’d want to be in that mix. To achieve this we first taught a machine learning system to understand what surfing looks like. We do this by labelling thousands of images of people surfing from years of camera data at hundreds of locations. Next we teach the system to understand how these people surfing move through time — this gives it an understanding of how to join the video frames together into a series of events. Once we understand this we know a great deal more about what’s going in the water: number of rides per hour, average ride length, average ride duration and the details of every single wave surfed.

The view from our office window is also the view from our HB Pier camera, which means field testing is a perk of the job.

So now we have a camera that, instead of just relaying real-time video, can show you the last five waves surfed. It can alert you if there are more waves being surfed than normal or if the average ride length hits your requirements. It’s also a camera that can deliver to you all the waves surfed during your session — an intelligent rewind of every swell you surf.

Our system works at a range of waves. Our Pipeline camera helped us test our ability to track surfers in the barrel and even, occasionally, the air.

The same camera is also keeping checks how many people there are in the water, where they’re sitting, how they’re being moved around by the currents and even where and how waves are breaking. All now hard data that can be used to make sure you’re in the right place at the right time.

One concept for a next-generation camera surf check based on our intelligent camera data.

Real-time intelligent surf cameras will be live on surfline.com soon.

Any questions about our camera technology please fire away in the comments below, or find me on twitter.

--

--

Ben Freeston
Surfline Labs

VP of data science at Surfline + Magicseaweed. Checking charts and chasing waves.