Re:High-res image reconstruction with latent diffusion model...
by dwringer on Friday March 03, 2023 @03:59PM
(#35014915)
> The problem is that it's impossible to know what is in the fMRI data and what is hallucinated by the reconstruction.
The same problem exists when prompting stable diffusion normally. However, the real interesting thing would be if subjects could be trained to vividly imagine looking at something and see if the interface could reconstruct it from that. Though perhaps a stepping stone, it does not necessarily seem trivial to jump from this to that.
The same problem exists when prompting stable diffusion normally. However, the real interesting thing would be if subjects could be trained to vividly imagine looking at something and see if the interface could reconstruct it from that. Though perhaps a stepping stone, it does not necessarily seem trivial to jump from this to that.