Hillel's Tech Corner: Sightbit's AI offers lifeguards internal binoculars

Sightbit is a smart lifeguard system that prevents drowning, making swim sites safer and more efficient to manage.

Sightbit (photo credit: Courtesy)
(photo credit: Courtesy)
Summer officially starts on June 20th, but in Israel that summer weather is clearly already here, and then some. With it, comes the usual round of summer activities – vacations, barbecues and trips to the beach. However, those trips to the beach aren’t as harmless as they appear. According to the World Health Organization, drowning is the third-leading cause of unintentional injury-related death, and is one of the top 5 causes of death for people aged 1-14 years for 48 of 85 countries with data meeting inclusion criteria.
Sure, there are lifeguards on duty in public beaches, but how much can they truly be expected to monitor? In addition to keeping an eye on hundreds of scattered people, there are also hazardous areas, rip currents and unforeseen circumstances (such as the inability to swim because of water temperature or muscle cramps) that should be factored to prevent potential drowning incidents. At the end of the day, lifeguards are still humans that are prone to error. Their work is far more complicated than what popular television programs would have you believe. Children sometimes get lost as they wander alone. Great swimmers get carried away in an instant by strong currents. People swim out too far, or sometimes are unable to swim after falling out of a boat.
After hearing about one too many drowning incidents, and given all the tech to which I am exposed, I asked myself whether there is some way to harness the power of artificial intelligence, which seems to be powering all our tech today, to save lives and prevent drowning. Well, it turns out there is a solution out there and what do you know? It’s Israeli.
Meet Sightbit.
Sightbiit was founded just over a year ago when co-founders Netanel and Adam were at the beach. The way he tells it, he saw one of the lifeguards pull out a pair of binoculars to get a better look at the swimmers. So we’re talking the same equipment used by 19th century birdwatchers.
Adam couldn’t believe that in a world in which a surgeon can operate on a person on the other side of the globe using remote-controlled robots, lifeguards rely on eyesight – and sometimes binoculars – to keep millions of people safe at the beach.
The company, which has offices in Beersheba, is a smart lifeguard system that prevents drowning, making swim sites safer and more efficient to manage. It’s a camera-based system that instantly warns of danger to swimmers. It can pretty much use any camera off the shelf because the magic is done with its software.
Alerts allow human lifeguards and swim agencies to respond immediately and save lives. Their AI software predicts danger in advance, so that swim agencies are always prepared. Sightbit is an early-stage company that is piloting its technology in Israel, in Palmahim Beach in collaboration with the Israel Nature and Parks Authority. Their four-person founder team is supported by Cactus Capital, Ben-Gurion University of the Negev’s venture capital investment fund, a venture capital investment fund initiated by Yazamut 360, BGU’s entrepreneurship center.
Sightbit’s algorithms enable autonomous or semi-autonomous monitoring, oversight and management of aquatic sites. Their core capabilities are rooted in cutting-edge deep learning technology. Harnessing the latest advancements in computer vision, they utilize convolutional neural network architectures for object detection, feature extraction and real-time and predictive analytics.
Thanks to the AI component of Sightbit, all swimmers (and children at risk) within range are monitored all the time with pinpoint accuracy. Areas become significantly more guarded, without an increase in cost for swim agencies. Management teams will also gain insights that go far beyond what they see with their own eyes, such as risk analytics, assessment of crowding, manpower management and sea-state forecasting.
Sightbit CEO Netanel Eliav, previously a business and crisis consultant for major clientele, including the Prime Minister’s Office, focuses on finance, business development and strategic vision. CTO Gadi Kovler is a biomedical engineer with expertise in artificial intelligence and signal processing, who currently focuses on R&D, AI and machine learning. COO Adam Bismut, an experienced manager with a track record of building results-oriented organizations, focuses on operations. CMO Minna Shezaf is a marketing specialist and resource-development professional who also manages  client relations.
In a short time, the company partnered with Cactus Capital, fresh fund, BGU, The YU Innovation Lab, Nvidia Inception Program and Amazon Web Services.
The Sightbit team plans to expand internationally and has been invited to conduct additional pilots at beaches around the world, from the US to Scandinavia to the Balkans.
In the US alone, there are more than 5,000 coastal beaches, 4,000 coastal cities, 70,000 lakeside beaches, 12,000 marinas and 300,000 public pools that can directly benefit from Sightbit’s technology.
I think we can all agree that humans are not built for visual monitoring, especially at a large scale and on an ongoing basis. Even the best lifeguards cannot track all swimmers all the time. And let’s be honest, we all know that lifeguards cannot be everywhere at once, which results in unguarded swimming areas, and difficulty predicting crowding and danger.
Here is how the platform works. All video feedback from the cameras gets converged into one stream of data via a router into the company’s main GPU/CPU. From then on, their algorithms analyze the video. The output goes to screens, which in this case are to be reviewed by the lifeguard. The lifeguards can give input and control the UI. It’s a two-way communication. Meaning, the lifeguard controls what alerts he sees. Different switch buttons include: children (in dangerous areas), vessels (boats, jet skis, etc.), and rip currents. The lifeguards can choose to turn on these features in order to get warnings related to them.
Once the lifeguard clicks something, that gets sent back to the CPU. The team then also gets alerts. When the alerts pop up, the lifeguard can click on it and then the program zooms in for him and sees exactly where the alert is. As is, today, if a lifeguard wants to see a threat more clearly, he must use binoculars because you can’t see clearly enough with the naked eye alone.
The platform also displays statistics on a dashboard that shows weather conditions such as temperature, humidity, date, time, wind speed, population count, and risk level, and a recommendation for how many lifeguards are needed.
While Sightbit is at the beginning of its journey, it has accomplished an incredible amount with very limited resources. If my career in tech and business has taught me anything, it is that resourcefulness is the sign of a promising venture. Sightbit has a strong team, a critical mission and, from where I am sitting, seems to be making all the right moves.
I, for one, hope they continue down this path. Many lives depend on it.