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

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Careil, Marlène, Benchetrit, Yohann, King, Jean-Rémi
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866916746701570048
author Careil, Marlène
Benchetrit, Yohann
King, Jean-Rémi
author_facet Careil, Marlène
Benchetrit, Yohann
King, Jean-Rémi
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2505_14556
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dynadiff: Single-stage Decoding of Images from Continuously Evolving fMRI
Careil, Marlène
Benchetrit, Yohann
King, Jean-Rémi
Computer Vision and Pattern Recognition
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.
title Dynadiff: Single-stage Decoding of Images from Continuously Evolving fMRI
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.14556