Osaka University, Japan, scientists discovered a new method of converting neural imaging into visuals. By using magnetic resonance imaging (MRI) with AI technology, scientists created visual images from human brain activity. The researchers used Stable Diffusion technology, a cost-effective platform that trains itself using images already stored on the internet, adapting to develop visuals that do not physically exist.
From The National Pulse:
To produce their images, the Japanese researchers followed a two-stage process. First, they decoded a visual image from the MRI signals of their test subjects. Then they used the MRI signals to “decode latent text representations” which could be fed into the Stable Diffusion platform, like prompts, to enhance the quality of the initial visual images retrieved. ... You don’t have to be a fan of the Terminator or Matrix franchises, though, to see that the worrying potential uses of AI extend well beyond cheating university assessments or forcing mediocre artists out of business. The website Popular Mechnanics recently reported, for instance, that US Air Force drones can now recognise individual people’s faces from high in the air, using AI. The firm that has created the technology, Seattle-based RealNetworks, says the drones can use the technology to distinguish friend from foe, and that the software can be applied for rescue missions, patrolling and “domestic search operations”. An Israeli firm is working on similar technology that will help drones find the perfect angle for facial recognition.
