TY - GEN
T1 - Cross-Subject Mind Decoding from Inaccurate Representations
AU - Xu, Yangyang
AU - Liu, Bangzhen
AU - Shao, Wenqi
AU - Du, Yong
AU - He, Shengfeng
AU - Zhu, Tingting
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - 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.
AB - 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.
UR - https://www.scopus.com/pages/publications/105044148329
U2 - 10.1109/ICCV51701.2025.01398
DO - 10.1109/ICCV51701.2025.01398
M3 - 会议稿件
AN - SCOPUS:105044148329
T3 - Proceedings of the IEEE International Conference on Computer Vision
SP - 15066
EP - 15075
BT - Proceedings - 2025 IEEE/CVF International Conference on Computer Vision, ICCV 2025
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2025 IEEE/CVF International Conference on Computer Vision, ICCV 2025
Y2 - 19 October 2025 through 23 October 2025
ER -