Reconstructing visual experiences from human brain activity offers a unique way to understand how the brain represents the world, and to interpret the connection between computer vision models and our visual system. While deep generative models have recently been employed for this task, reconstructing realistic images with high semantic fidelity is still a challenging problem. Here, we propose a new method based on a diffusion model (DM) to reconstruct images from human brain activity obtained via functional magnetic resonance imaging (fMRI). More specifically, we rely on a latent diffusion model (LDM) termed Stable Diffusion. This model reduces the computational cost of DMs, while preserving their high generative performance. We also characterize the inner mechanisms of the LDM by studying how its different components (such as the latent vector of image Z, conditioning inputs C, and different elements of the denoising U-Net) relate to distinct brain functions. We show that our proposed method can reconstruct high-resolution images with high fidelity in straight-forward fashion, without the need for any additional training and fine-tuning of complex deep-learning models. We also provide a quantitative interpretation of different LDM components from a neuroscientific perspective. Overall, our study proposes a promising method for reconstructing images from human brain activity, and provides a new framework for understanding DMs. Please check out our webpage at . ### Competing Interest Statement The authors have declared no competing interest.

Soon “You are arrested. You have the right to remain silent, however, everything you think can be used against you” 🥶

Presumably this could go the other way eventually as well to construct sensations directly in the brain. That’ll be a whole new level of VR. You could potentially share mind states as well using this kind of tech. You’d encode an experience from one brain and then decode in another.

Create a post

Subscribe to see new publications and popular science coverage of current research on your homepage

  • 0 users online
  • 1 user / day
  • 4 users / week
  • 16 users / month
  • 120 users / 6 months
  • 3.13K subscribers
  • 1.06K Posts
  • Modlog