Decoding images from the human brain
Today, the journal Science published a scientific paper ( pdf ) describing a new method for the automatic reconstruction of images from the human brain. The authors of the scientific work believe that their method is more accurate than previous works in this area: alphabet characters and simple contrasting figures are well recognized.
So far, the decoder only recognizes pictures that a person actually sees at the moment, but in the future it should work on imaginary images. In this case, it will become possible, for example, to type text with eyes closed.
Image recognition is carried out by the probabilistic method, using the Bayesian classifier. In this case, the image decoder does not need training.
Functional magnetic resonance imaging (fMRI), a type of magnetic resonance imaging that is used to measure changes in blood flow, is used to capture the signal. There is a slight delay between neural activity and changes in blood flow, which is clearly seen in the demo video.
Anyone can try the image recognition program: it is posted on Github .
Magnetic resonance scan data and the converter can be taken here .
The result of the decoder is shown in the screenshots.
Science article