A system developed in Japan can describe a picture someone is viewing, using scans of their brain. Algorithms have recently become pretty good at generating captions when shown an image. Now it seems a similar technique can be used to generate captions when an artificial intelligence is shown scans of a person's brain, taken while they were looking at an image. To generate a caption, the Al is given an image of a person's brain, taken with an fMRI scanner. The fMRI scanner shows the surges in blood flow that correspond with activity, so the different parts of the brain involved in processing the image light up on the scan. From this, the Al then produces a caption based on what it thinks the person was viewing
© 2001-2024 Fundación Dialnet · Todos los derechos reservados