Scientists convert brain waves to image
- Thread starter FreddyE
- Start date
Um... I don't think that's a real article... did you see some of the other ones from the original source website? Fear detector being developed to sniff out terrorists? It literally "smells" fear off the terrorists? Not to mention, did you read the wording in the article? No way that's real. Especially when it's only been so recently that they could get up and down out of brain waves. Image is WAY too complicated. Binary is what they get out of brain waves. To get a color? The color red in binary is: 111111110000000000000000. That's pure red. Not to mention all the shades. look at all those digits. That's just to get the COLOR... then you'd need to note it's position. And how many pixels are in this image they are pulling from your head? There's no way any of our computers could pull images as detailed as their examples, if they could even pull A COLOR that you were thinking of, I'd be impressed.
Don't worry, nothing to be creeped out by. It's not anywhere close to being possible yet.
Don't worry, nothing to be creeped out by. It's not anywhere close to being possible yet.
I struggle to believe a computer could pull up pictures like that. The brain doesn't think in full resolution bitmap pictures -- only a tiny part of brain activity is going to be in that form, which is what comes straight from the optic nerve. And even then, it's not in full colour or focus -- Central vision is in detail in tri-colour. Peripheral vision is imprecise and typically greyscale. The overwhelming majority of data they'd sense would be a mishmash of conceptual interpretation, and heavily filtered (frame problem anyone?), with mental focus leaping around all over the place.
Hmmm... here's the PI's bio:
http://neuroscience.berkeley.edu/users/users_profile.php?id=12
He's a professor of psychology, but working at Berkeley in Vision Science.
http://vision.berkeley.edu/VSP/
Apparently, the information was from a March 5, 2008 Nature article
http://www.nature.com/news/2008/080305/full/news.2008.650.html
a bit more of the article is available here:
http://vision.berkeley.edu/VSP/content/news_events/news.html
Interesting stuff, and real science, none of that that "lets' pretend" crap.
http://neuroscience.berkeley.edu/users/users_profile.php?id=12
He's a professor of psychology, but working at Berkeley in Vision Science.
http://vision.berkeley.edu/VSP/
Apparently, the information was from a March 5, 2008 Nature article
http://www.nature.com/news/2008/080305/full/news.2008.650.html
a bit more of the article is available here:
http://vision.berkeley.edu/VSP/content/news_events/news.html
Interesting stuff, and real science, none of that that "lets' pretend" crap.
I've heard of this about a year ago.
Personally, I think those developing this are trying to come up with any excuse they can think of to justify this development.
It is truly astounding that the fact that this would almost inevitably be used as an interrogation tool would not occur to those developing it.
CuttingEdge100
Personally, I think those developing this are trying to come up with any excuse they can think of to justify this development.
It is truly astounding that the fact that this would almost inevitably be used as an interrogation tool would not occur to those developing it.
CuttingEdge100
Like so many technological developments, it could be used for beneficial, nefarious, or just plain foolish purposes.
Actually, now that I read both these articles thoroughly it would seem they are planning to develop it into a crime solving tool.
First it will be used to help eye-witnesses recall a crime better, than, if not right away, once everybody's comfortable with that it will be used as an interrogation tool.
CuttingEdge100
First it will be used to help eye-witnesses recall a crime better, than, if not right away, once everybody's comfortable with that it will be used as an interrogation tool.
CuttingEdge100
Ah, a simple generative model, it sounds like. Impressive, but hardly ground-breaking. The technology has been around for years, the only difference is what inputs they're choosing to give it.
It's like if I held my left hand up when I saw yellow, and my right hand up when I saw red. Then I hold both hands up, and, viola, you predict I'm seeing orange and everyone hails you as a genius.
If I'm right, all they can really do is generate images which are a "mixture" of one or more of the training images. The weights of each image are really all that's being discerned; how similar the brain's response is to what it did for each of them. Of course, what exactly it means to mix images isn't a well-defined question, so that could be an interesting algorithm.
Last edited:
Like so many technological developments, this will be mainly used as a method for creating pornography. Depending on your value-set that could be considered beneficial, nefarious, or just plain foolish.
That was my first thought. If I imagine Deanna Troi making out with Beverly Crusher this machine will make a picture of it? AWESOME.
Like so many technological developments, this will be mainly used as a method for creating pornography. Depending on your value-set that could be considered beneficial, nefarious, or just plain foolish.
Beneficial for those in the business, nefarious to some with a particular set of morals, and foolish? See this link.
If we could construct a brain the size of Idaho, Oregon, and Washington state combined, would we get a massively larger picture from the device?
Or the ones that use the best drugs, which means Hollywood would still be #1 when it comes to coming up with crazy stuff.
Similar threads
