Japan's First Artificial Intelligence Recreates Images From Brain Signals


(MENAFN- AzerNews) Japanese specialists were able for the first time in the world to develop a technology that uses generative artificial intelligence (AI) to create images based on the processing of human brain signals. Kyodo agency reported about it, Azernews reports.

During the experiment, a research team from the National Institute of Quantum Science and Technology recorded brain signals at the time of viewing human images of objects and landscapes. The participants of the experiment were shown about 1,200 different images. Brain activity was analysed using functional magnetic resonance imaging (fMRI). The same images were uploaded to an AI database, which was tasked with learning how to correlate them with brain signals.

Previous studies have already shown that images seen by humans can be reconstructed from brain signals processed using functional MRI. However, only a limited range of images could be reproduced in this way.

Now Japanese scientists, based on previous achievements, have developed a programme that converts brain signals into numerical values, on the basis of which already trained artificial intelligence recreates images.

The image of a leopard with clearly drawn mouth, ears and spotted coat colouring turned out to be the closest to reality.

Scientists believe that this technology can be used in the development of communication devices, as well as to understand the mechanisms of hallucinations and dreams.

MENAFN16122023000195011045ID1107607304


AzerNews

Legal Disclaimer:
MENAFN provides the information “as is” without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the provider above.