11–20 of 97 posts
If I were a billionaire, one of the projects I'd do would be to apply this sort of technology at scale, upon volunteers who were witness to poorly recorded historic events. WWII, Beatles concerts, JFK assassination, 9/11... all of them have been recorded from angles/perspectives that exist only in peoples' memories, and the longer we wait, the more of those memories will be lost.
https://www.biorxiv.org/content/biorxiv/early/2017/12/28/240...
When we build neural networks to decode our own neural networks...
Wow, I'm calling a hoax. They produce images of dubious value, and for whatever reason, contaminate the results by mutating them with the Deep Dream stylization. This is as bad as the dead salmon results. EDIT: You can flag all you want, but look at this playback of the reconstruction: https://video.twimg.com/tweet_video/DSrWxhSVQAE2Xix.mp4 There's nothing useful happening here.
https://github.com/KamitaniLab/DeepImageReconstruction
Not a hoax.
Your comparison to Deep Dream is quite apt, for good reason that you clearly don't realize-
Deep Dream is an attempt to "enhance patterns in images via algorithmic pareidolia"- Pareidolia being a "psychological phenomenon in which the mind responds to a stimulus". So yes, this project is affected by a real life "deep dream".
It's interesting that they were not able to extract good images from the test when they just asked the subjects to imagine the pictures. I speculate that is because the vast majority of people are able to recognize images of things, but not produce those images.
For a simple example, see this article about people trying to draw a bicycle. Before clicking, try to draw a 2D representation of a bicycle -- the wheels, the frame, pedals, handlebars, etc. You'd be surprised how difficult it is. https://road.cc/content/blog/90885-science-cycology-can-you-...
I would be very interested to see them repeat the experiment on e.g. artists who specialize in photorealism, or asking an expert on a very particular thing to imagine that particular thing (e.g. ask a violinist to imagine a violin). I wonder if that would yield more recognizable results than asking a random grad student to visualize a lion's face.
Next: All you need to do to make the next billion $ movie is your imagination, literally (and a 100m$ helium cooled MRI machine, and 25$ worth of AWS credits).
MRI is crazy expensive, but you only really need $4M for the scanner, so doable for indie rom-come as well as blockbusters!
Among useful applications I would imagine, for example, the fast image-based information retrieval - e.g. useful for recommendation systems (in other words - Netflix could invest in a 'imagine anything and we find you a movie like that').
Acknowledging that I know nothing about psychology or neuroscience... It's interesting that they were not able to extract good images from the test when they just asked the subjects to imagine the pictures. I speculate that is because the vast majority of people are able to recognize images of things, but not produce those images. For a simple example, see this article about people trying to draw a bicycle. Before cl…
Hmm. Do different humans produce relatively the same brain images for a given scene? If not, I’d imagine training this would be a real PITA.