AI Computer Predicts Pictures Via Human Thoughts

Original article was published by Mindstick on Artificial Intelligence on Medium

AI Computer Predicts Pictures Via Human Thoughts

Techniques of Artificial Intelligence are not only affecting the brain but also trying to control it. In a similar effort, researchers at the University of Helsinki have developed a technology in which the computer reads the brain’s mind and guesses their visual perception. In this way, the computer conceives in a way the human being is thinking about.

In this way, by extracting the pictures, the computer can gain a new kind of information. Like fictional photos that have never been seen. This technology is based on a new type of interface created between the brain and the computer. Even before this, interfaces between brain and computer were made. But they could only do one-way communication. To move the cursor or to write the letter spelling.

This study research has been published in the journal Scientific Reports this month. Researchers call this technique neuroadaptive generative modeling. The study included 31 participants to measure the efficacy of the technique.

In this study, the researchers showed the participants hundreds of photographs made with the help of artificial intelligence. These pictures were of different types of people while EEG was being recorded while showing the picture.

The participants were asked to look at the special things while looking at the pictures like how old are the faces, are they smiling, etc. During quick viewing of these photographs, the participants were filled in the EEG Neural Network which ascertained whether the picture that the participants were seeing, their mind is able to conclude in the same way that they are looking for.

On the basis of this information, Neural Network guessed the photos that the participants were thinking about. Eventually, the participants of the photographs made by the computer were evaluated and these pictures were shown to be very similar to the pictures they had thought. Experiment accuracy turned out to be 83 percent.

“This technique combines natural human responses and the computer’s ability to extract new information,” said Tucca Rotsalo, a research fellow at the Academy of Finland at the University of Helsinki.

Researchers say that taking photographs of the face of a human being is only one example of efficient use of this technique. Human creativity can also be increased by this. Many kinds of unconscious attitudes can also be revealed from this, which are not usually exposed.

Technology does not identify ideas, but rather reacts to things related to parts of the brain. This technique can be used in the social, cognitive, and emotional processes of the brain.