dynadiff / README.md
ybenchetrit's picture
Super-squash branch 'main' using huggingface_hub
0c2ed76 verified
metadata
license: other
license_name: fair-noncommercial-research-license
license_link: https://huggingface.co/facebook/dynadiff/blob/main/LICENSE
extra_gated_prompt: >-
  Please be sure to provide your full legal name, date of birth, and full
  organization name with all corporate identifiers. Avoid the use of acronyms
  and special characters. Failure to follow these instructions may prevent you
  from accessing this model and others on Hugging Face. You will not have the
  ability to edit this form after submission, so please ensure all information
  is accurate.
extra_gated_fields:
  First Name: text
  Last Name: text
  Date of birth: date_picker
  Country: country
  Affiliation: text
  I accept the terms and conditions: checkbox
  geo: ip_location
language:
  - en
tags:
  - neural-decoding
  - fMRI
  - brain-to-image

Dynadiff: Single-stage Decoding of Images from Continuously Evolving fMRI

This repository contains the model weights for our paper: "Single-stage Decoding of Images from Continuously Evolving fMRI".

Abstract

Brain-to-image decoding has been recently propelled by the progress in generative AI models and the availability of large ultra-high field functional Magnetic Resonance Imaging (fMRI). However, current approaches depend on complicated multi-stage pipelines and preprocessing steps that typically collapse the temporal dimension of brain recordings, thereby limiting time-resolved brain decoders. Here, we introduce Dynadiff (Dynamic Neural Activity Diffusion for Image Reconstruction), a new single-stage diffusion model designed for reconstructing images from dynamically evolving fMRI recordings. Our approach offers three main contributions. First, Dynadiff simplifies training as compared to existing approaches. Second, our model outperforms state-of-the-art models on time-resolved fMRI signals, especially on high-level semantic image reconstruction metrics, while remaining competitive on preprocessed fMRI data that collapse time. Third, this approach allows a precise characterization of the evolution of image representations in brain activity. Overall, this work lays the foundation for time-resolved brain-to-image decoding.

To run the model, see the readme here: https://github.com/facebookresearch/dynadiff