AI vision has a few capabilities that human vision does not have (the capability to spot diseases, for example), and its capability to go beyond human-vision capability is increasing and finding broader application. For example, many AI‑vision systems can now reconstruct objects that a foreground obstruction is hiding from their visual field.
Systems can also reconstruct what an image of a phenomenon might have looked like in the past or uncover specific details (such as concealed weapons) in real-time video streams at a rate that far exceeds the rate that is possible with human vision. As with all modern AI applications, AI capable of going beyond human vision can be prone to failure, but its utility likely outweighs its drawbacks.
Abstracts That Inspired This Pattern
"Artificial intelligence can use your brainwaves to see around corners. The technique, called 'ghost imaging,' can reconstruct the basic details of objects hidden from view by analysing how the brain processes barely visible reflections on a wall."
Researchers at Cardiff University have developed a way to use machine learning to explore the past of a supernova. After undergoing training on hundreds of examples of supernovas, an algorithm "can construct whole artificial spectra based on only a few data points" from supernovas that researchers observed previously, and "filling in the gaps for these existing data points then allows a spectrum to be constructed for any past explosion up to around 200 days past the explosion."
US-based Evolv Technology has developed screening systems that search for weapons. The smart systems, which can scan about 60 people per minute, create images via a light-emission technique and then use AI to examine those images. With $400 million in funding, the company is gaining popularity and supplying numerous public venues with its technology.