News classification
Contact us
- Add: No. 9, North Fourth Ring Road, Haidian District, Beijing. It mainly includes face recognition, living detection, ID card recognition, bank card recognition, business card recognition, license plate recognition, OCR recognition, and intelligent recognition technology.
- Tel: 13146317170 廖经理
- Fax:
- Email: 398017534@qq.com
Artificial intelligence AI reading there
Artificial intelligence AI reading there
In the Mel Gibson film "overheard the woman heart", the protagonist Nick has the uncanny ability to read minds, presumably so many people envy. Now, with the help of AI technology, this magical ability is becoming a reality.
Researchers at Kyoto University in Japan Kamitani lab recently invented a AI machine, we study electronic signals in the brain of the AI, according to the magnetic resonance imaging (fMRI) scan information to accurately calculate the image one is reading, or is thinking about things. This technology opens a door for some future scenarios, such as the role played in black mirror, anyone can record and playback their memories.
According to the daily mail, the discovery was made by researchers at the Kamitani laboratory at the Kyoto University and led by Professor Yukiyasu Kamitani. Experts make use of neural networks to make images based on the information scanned by magnetic resonance imaging (fMRI). Magnetic resonance images can detect changes in blood flow to analyze electrical activity.
Using this data, the machine could rebuild owls, planes, colored glass windows and red mail boxes, after 3 volunteers stared at the pictures.
It also made some pictures of objects, including squares, crosses, goldfish, swans, leopards and bowling balls, which were also imagined by participants.
According to the research group, this breakthrough opens the "unique window of our inner world", despite the differences between people.
In theory, this technique can be used to produce images of daydreams, memories and other mental images. In addition, it can also help permanent vegetative patients communicate with their loved ones.
This breakthrough depends on the neural network, which tries to simulate the way the brain works to learn. The object of the study shows a natural image on an online image database ImageNet, which contains 150 categories.
Then, the trained decoder is used to predict the visual features of the object - even those that are not used in the brain scan training. When the same image was displayed, the researchers found that the brain activity patterns of the human body were very similar to the models that simulated neurons in the neural network.
Now, more and more progress on the AI prediction of human thinking, though they are still in their infancy, decoding some human visual see some simple image basically, but we can still see clearly, in AI in the prediction of brain activity, in a step by step forward.
Researchers at Kyoto University in Japan Kamitani lab recently invented a AI machine, we study electronic signals in the brain of the AI, according to the magnetic resonance imaging (fMRI) scan information to accurately calculate the image one is reading, or is thinking about things. This technology opens a door for some future scenarios, such as the role played in black mirror, anyone can record and playback their memories.
According to the daily mail, the discovery was made by researchers at the Kamitani laboratory at the Kyoto University and led by Professor Yukiyasu Kamitani. Experts make use of neural networks to make images based on the information scanned by magnetic resonance imaging (fMRI). Magnetic resonance images can detect changes in blood flow to analyze electrical activity.
Using this data, the machine could rebuild owls, planes, colored glass windows and red mail boxes, after 3 volunteers stared at the pictures.
It also made some pictures of objects, including squares, crosses, goldfish, swans, leopards and bowling balls, which were also imagined by participants.
According to the research group, this breakthrough opens the "unique window of our inner world", despite the differences between people.
In theory, this technique can be used to produce images of daydreams, memories and other mental images. In addition, it can also help permanent vegetative patients communicate with their loved ones.
This breakthrough depends on the neural network, which tries to simulate the way the brain works to learn. The object of the study shows a natural image on an online image database ImageNet, which contains 150 categories.
Then, the trained decoder is used to predict the visual features of the object - even those that are not used in the brain scan training. When the same image was displayed, the researchers found that the brain activity patterns of the human body were very similar to the models that simulated neurons in the neural network.
Now, more and more progress on the AI prediction of human thinking, though they are still in their infancy, decoding some human visual see some simple image basically, but we can still see clearly, in AI in the prediction of brain activity, in a step by step forward.