Several years ago, people believe the mind reading is a science fiction or a spiritual thing people who meditate can do. But recently, there are some research findings that prove mind reading is possible. Researchers have tried it several reconstruction outputs including, characters, face, sound, images, etc. Because of the development of the machine learning and deep learning methodologies such as Deep Neural Networks (DNN), Generative Adversarial Networks (GAN), etc., these reconstruction methods are getting better and better.
Prof. Yukiyasu Kamitani, a professor at the Kyoto University (KamitaniLab) and his team have been working on this subject in last decade. They have uploaded a video of the demonstration of there recent results which is reconstructing the imaginations of the images using fMRI scans. Even though reconstructed images are not so clear still we can get some idea from them. Because of the improvement of deep learning research field, we can expect better results in the future.
Related research article can be found here. https://doi.org/10.1101/240317
Another interesting article has been published by Alan S. Cowen et.al. which is about the reconstruction of the faces from the evoked brain activity. They have used fMRI scans as input data then find the best features to reconstruct the face using Principal Component Analysis (PCA) and finally, reconstruct the image using machine learning method.
Research article: 10.1016/j.neuroimage.2014.03.018
Not only images but, some scientists developed a system to decode brain waves into sound. An electronic voice was spoken to the patients and by analyzing brain waves system provides what they had heard. Following video (Credits: https://www.telegraph.co.uk/news/science/science-news/9051909/Mind-reading-device-could-become-reality.html) shows the demonstration of the developed system.
UC Berkeley research team have developed a method to reconstruct people’s dynamic visual experience using Hollywood movie trailers. They also have used fMRI brain scans to do that. Prof. Jack L. Gallant is explaining how they did that and the results of the research in below video.
More research papers
- Identifying natural images from human brain activity
- Bayesian Reconstruction of Natural Images from Human Brain Activity
- Visual Image Reconstruction from Human Brain Activity using a Combination of Multiscale Local Image Decoders Author links open overlay panel
- Sharing deep generative representation for perceived image reconstruction from human brain activity
- Linear reconstruction of perceived images from human brain activity
More related videos