The researchers focused on the visual cortex, the ‘seeing’ part of the brain, which they analyzed through hundreds of fMRI scans of people looking at a single handwritten number or letter, along with the image they were looking at. Using a deep generative multiview model (DGMM) algorithm, they could reconstruct the original images from the fMRI scans.Decoding more complex pictures will need a more sophisticated approach, involving AI for reconstruction.
CAS news release, May 11, 2017