The Eye of AI #2

Computer vision is a truly transformational technology for the retail industry, but it has applications that touch nearly every aspect of our lives. This is our weekly summary of what happens when cameras, computers and great ideas get together.

Baaaaah-bum

The University of California Santa Barbara's Benioff Ocean Science Laboratory (BOSL) has launched a project to identify aquatic predators using drone footage. The aim is to keep beaches safer for swimmers and surfers, while also monitoring changing shark habits due to climate change making waters warmer. It could also help police chiefs estimate boat sizes before putting to sea.

Olympic Talent Spotting

AI is being used in many new ways at the Paris Olympics. Pushing faster, higher, stronger beyond the current olympiad, Intel has developed 3D athlete tracking technology which is being used to identify future Olympic talent in Senegal. Democratising access to this AI is a fraught issue, but making sure everyone can use this type of tech is vital to ensuring the global playing field remains level and sport stays truly competitive.

An Eye on Natural Depth Perception

Nature trades off speed of perception against the benefits of seeing in 3D. Humans can experience glorious landscapes, but can't dodge a flying golf ball for instance. Conversely, it's hard to swat a fly because their eyes are adapted to quickly identify movement within their field of vision. They're terrible at depth-perception though, which is why they bump into everything when they fly away. Researchers at the University of Virginia School of Engineering and Applied Science turned their gaze to the Preying Mantis - a creature that is uniquely adapted to fast perception and appreciating space. Special lenses based on mantis eyes mean less processing is required to judge motion and depth, key in preventing autonomous vehicle accidents, and perhaps leading to robotic miniature kung-fu masters - or something...

Smarter Spraying

There aren't many people who like the idea of spraying chemicals on crops. While they definitely help farmers produce reliable yields, pesticides are expensive and have knock-on effects throughout ecosystems. Which is why it makes sense to know that you're spraying the right thing in the right place. So fitting your existing machinery with cameras that identify the presence (and type) of weeds, and only targeting affected areas is a big step in the right direction.

Say A-Haa!

Dentists are notorious for constantly looking down in the mouth, so this should be something to lift their spirits. Pearl, based in the bright-smile state of California, aims to make visits to the dentist faster and more beneficial by using computer vision to analyse radiographs in real-time to highlight oral health issues.

That's everything for this week. Please keep an eye on the SAI Group blog for everything that we're talking about, including a new article this week about shrinkage at the self-checkout.

Got some cool tech to share? Whether its your own project, or just something you saw and thought "I want people to know about this!", let us know about it and we'll include it in upcoming editions.