Using electrical signals from human brains, new software can perform automated image editing

Brain responses are used as supervisory signals for semantic image processing. Credit: Tuukka Ruotsalo et al

Soon, computers could sense that users have a problem and come to the rescue. This is one of the possible implications of new research at the University of Copenhagen and the University of Helsinki.

“We can have a computer edit images based entirely on thoughts generated by human subjects. The computer has absolutely no prior information about what functions to edit or how. No one has ever done this before,” said associate professor Tuukka Ruotsalo , Department of Computer Science, University of Copenhagen.

The results are presented in a paper accepted for publication at the CVPR 2022 (Computer Vision and Pattern Recognition).

Brain activity as the only input

In the underlying study, 30 participants were fitted with hoods with EEG electrodes that map electrical brain signals. All participants were given the same 200 facial images to look at. They were also given a series of tasks such as searching for female faces, searching for elderlylooking for blond hair, etc.

The participants performed no actions and looked at the images briefly – 0.5 second for each image. based on their brain activity, the machine first mapped the given preference and then edited the images accordingly. So if the job was to find older people, the computer would adjust the portraits of the younger people, making them look older. And if the job was to search for a particular hair color, all images would be that color.

“Notably, the computer had no knowledge of facial recognition and had no idea of ​​gender, hair color or other relevant functions. Yet it only edited the function in question, leaving other facial features unchanged’, says PhD student Keith Davis, University of Helsinki.

Some might argue that there is already a lot of software that can manipulate facial features. That would miss the point, Keith Davis explains:

“All existing software has been trained before with labeled inputs. So if you want an app that can make people look older, you run those thousands of portraits and tell the computer which ones are young and which ones are old. The subjects’ brain activity was the only one.” input This is a whole new paradigm in artificial intelligence– using the human brain directly as an input source.”

Potential applications in medicine

A possible application could be in medicine: “Doctors already use artificial intelligence in the interpretation of scan images. However, mistakes do happen. After all, the doctors are only assisted by the images, but make the decisions themselves. Perhaps certain features in the images are more often misinterpreted than others. Such patterns can be discovered through an application of our research,” says Tuukka Ruotsalo.

Another application can help certain groups of people with disabilities, for example by allowing a paralyzed person to operate his or her computer.

“That would be fantastic,” says Tuukka Ruotsalo. “However, that is not the focus of our research. We have a broad scope and want to improve machine learning in general. possible applications will be wide. In 10 or 20 years, for example, we won’t have to use a mouse or type commands to get our computer† Maybe we could just use mind control.”

Calls for policy regulation

However, the coin also has a flip side, according to Tuukka Ruotsalo: “Collecting individual brain signals does raise ethical issues. Acquiring this knowledge may be able to gain deep insight into someone’s preferences. We are already seeing some trends. ‘watches and similar devices that can record heartbeat etc., but are we sure no data is being generated that provides private companies with knowledge we wouldn’t want to share?”

“I see this as an important aspect of academic work. Our research shows what is possible, but we shouldn’t do things just because we can. This is an area that I believe should be regulated by guidelines and government policies. adjusted, private companies will continue.”

A computer predicts your thoughts and makes images based on that

More information:
Paper: … _CVPR_2022_paper.pdf


Quote: Using electrical signals from human brains, new software can perform automated image processing (2022, June 24), retrieved June 25, 2022 from computerized. html

This document is copyrighted. Other than fair dealing for personal study or research, nothing may be reproduced without written permission. The content is provided for informational purposes only.

#electrical #signals #human #brains #software #perform #automated #image #editing

Leave a Comment

Your email address will not be published. Required fields are marked *