And that's exactly why this is garbage research. Point deep dream at an interference pattern of pure noise, after training it on a set of target images.
Don't be surprised when it "finds" the image, pulled from raw static that contained no signal.
Don't be surprised when it "finds" the image, pulled from raw static that contained no signal.
> While our model was solely trained with natural images, our method successfully generalized the reconstruction to artificial shapes, indicating that our model indeed reconstructs or generates images from brain activity, not simply matches to exemplars.
If I am reading that correctly it is reconstructing images that the subjects had not seen before.