Cross-Subject Mind Decoding from Inaccurate Representations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xu, Yangyang, Liu, Bangzhen, Shao, Wenqi, Du, Yong, He, Shengfeng, Zhu, Tingting
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908466780569600
author Xu, Yangyang
Liu, Bangzhen
Shao, Wenqi
Du, Yong
He, Shengfeng
Zhu, Tingting
author_facet Xu, Yangyang
Liu, Bangzhen
Shao, Wenqi
Du, Yong
He, Shengfeng
Zhu, Tingting
contents Decoding stimulus images from fMRI signals has advanced with pre-trained generative models. However, existing methods struggle with cross-subject mappings due to cognitive variability and subject-specific differences. This challenge arises from sequential errors, where unidirectional mappings generate partially inaccurate representations that, when fed into diffusion models, accumulate errors and degrade reconstruction fidelity. To address this, we propose the Bidirectional Autoencoder Intertwining framework for accurate decoded representation prediction. Our approach unifies multiple subjects through a Subject Bias Modulation Module while leveraging bidirectional mapping to better capture data distributions for precise representation prediction. To further enhance fidelity when decoding representations into stimulus images, we introduce a Semantic Refinement Module to improve semantic representations and a Visual Coherence Module to mitigate the effects of inaccurate visual representations. Integrated with ControlNet and Stable Diffusion, our method outperforms state-of-the-art approaches on benchmark datasets in both qualitative and quantitative evaluations. Moreover, our framework exhibits strong adaptability to new subjects with minimal training samples.
format Preprint
id arxiv_https___arxiv_org_abs_2507_19071
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Cross-Subject Mind Decoding from Inaccurate Representations
Xu, Yangyang
Liu, Bangzhen
Shao, Wenqi
Du, Yong
He, Shengfeng
Zhu, Tingting
Computer Vision and Pattern Recognition
Decoding stimulus images from fMRI signals has advanced with pre-trained generative models. However, existing methods struggle with cross-subject mappings due to cognitive variability and subject-specific differences. This challenge arises from sequential errors, where unidirectional mappings generate partially inaccurate representations that, when fed into diffusion models, accumulate errors and degrade reconstruction fidelity. To address this, we propose the Bidirectional Autoencoder Intertwining framework for accurate decoded representation prediction. Our approach unifies multiple subjects through a Subject Bias Modulation Module while leveraging bidirectional mapping to better capture data distributions for precise representation prediction. To further enhance fidelity when decoding representations into stimulus images, we introduce a Semantic Refinement Module to improve semantic representations and a Visual Coherence Module to mitigate the effects of inaccurate visual representations. Integrated with ControlNet and Stable Diffusion, our method outperforms state-of-the-art approaches on benchmark datasets in both qualitative and quantitative evaluations. Moreover, our framework exhibits strong adaptability to new subjects with minimal training samples.
title Cross-Subject Mind Decoding from Inaccurate Representations
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.19071