Saved in:
Bibliographic Details
Main Authors: Tang, Jingyi, Jiang, Shuai, Su, Fei, Zhao, Zhicheng
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.07077
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911495466516480
author Tang, Jingyi
Jiang, Shuai
Su, Fei
Zhao, Zhicheng
author_facet Tang, Jingyi
Jiang, Shuai
Su, Fei
Zhao, Zhicheng
contents Visual decoding from electroencephalography (EEG) has emerged as a highly promising avenue for non-invasive brain-computer interfaces (BCIs). Existing EEG-based decoding methods predominantly align brain signals with the final-layer semantic embeddings of deep visual models. However, relying on these highly abstracted embeddings inevitably leads to severe cross-modal information mismatch. In this work, we introduce the concept of Neural Visibility and accordingly propose the EEG-Visible Layer Selection Strategy, aligning EEG signals with intermediate visual layers to minimize this mismatch. Furthermore, to accommodate the multi-stage nature of human visual processing, we propose a novel Hierarchically Complementary Fusion (HCF) framework that jointly integrates visual representations from different hierarchical levels. Extensive experiments demonstrate that our method achieves state-of-the-art performance, reaching an 84.6% accuracy (+21.4%) on zero-shot visual decoding on the THINGS-EEG dataset. Moreover, our method achieves up to a 129.8% performance gain across diverse EEG baselines, demonstrating its robust generalizability.
format Preprint
id arxiv_https___arxiv_org_abs_2603_07077
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Aligning What EEG Can See: Structural Representations for Brain-Vision Matching
Tang, Jingyi
Jiang, Shuai
Su, Fei
Zhao, Zhicheng
Computer Vision and Pattern Recognition
Visual decoding from electroencephalography (EEG) has emerged as a highly promising avenue for non-invasive brain-computer interfaces (BCIs). Existing EEG-based decoding methods predominantly align brain signals with the final-layer semantic embeddings of deep visual models. However, relying on these highly abstracted embeddings inevitably leads to severe cross-modal information mismatch. In this work, we introduce the concept of Neural Visibility and accordingly propose the EEG-Visible Layer Selection Strategy, aligning EEG signals with intermediate visual layers to minimize this mismatch. Furthermore, to accommodate the multi-stage nature of human visual processing, we propose a novel Hierarchically Complementary Fusion (HCF) framework that jointly integrates visual representations from different hierarchical levels. Extensive experiments demonstrate that our method achieves state-of-the-art performance, reaching an 84.6% accuracy (+21.4%) on zero-shot visual decoding on the THINGS-EEG dataset. Moreover, our method achieves up to a 129.8% performance gain across diverse EEG baselines, demonstrating its robust generalizability.
title Aligning What EEG Can See: Structural Representations for Brain-Vision Matching
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.07077