dorsal/arxiv
View SchemaLocal-Global Feature Fusion for Subject-Independent EEG Emotion Recognition
| Authors | Zheng Zhou, Isabella McEvoy, Camilo E. Valderrama |
|---|---|
| Categories | |
| ArXiv ID | 2601.08094vv1 |
| URL | https://arxiv.org/abs/2601.08094 |
| License | http://creativecommons.org/licenses/by/4.0/ |
Abstract
Subject-independent EEG emotion recognition is challenged by pronounced inter-subject variability and the difficulty of learning robust representations from short, noisy recordings. To address this, we propose a fusion framework that integrates (i) local, channel-wise descriptors and (ii) global, trial-level descriptors, improving cross-subject generalization on the SEED-VII dataset. Local representations are formed per channel by concatenating differential entropy with graph-theoretic features, while global representations summarize time-domain, spectral, and complexity characteristics at the trial level. These representations are fused in a dual-branch transformer with attention-based fusion and domain-adversarial regularization, with samples filtered by an intensity threshold. Experiments under a leave-one-subject-out protocol demonstrate that the proposed method consistently outperforms single-view and classical baselines, achieving approximately 40% mean accuracy in 7-class subject-independent emotion recognition. The code has been released at https://github.com/Danielz-z/LGF-EEG-Emotion.
{
"annotation_id": "cc0dc364-60fa-49cb-9e74-3373b4704f8e",
"date_created": "2026-02-17T05:53:16.068000Z",
"date_modified": "2026-02-17T05:53:16.068000Z",
"file_hash": "f978d7af05055aaa197a2b7c21abe25129f455907f20f84716356822c6235c3e",
"private": false,
"record": {
"abstract": "Subject-independent EEG emotion recognition is challenged by pronounced inter-subject variability and the difficulty of learning robust representations from short, noisy recordings. To address this, we propose a fusion framework that integrates (i) local, channel-wise descriptors and (ii) global, trial-level descriptors, improving cross-subject generalization on the SEED-VII dataset. Local representations are formed per channel by concatenating differential entropy with graph-theoretic features, while global representations summarize time-domain, spectral, and complexity characteristics at the trial level. These representations are fused in a dual-branch transformer with attention-based fusion and domain-adversarial regularization, with samples filtered by an intensity threshold. Experiments under a leave-one-subject-out protocol demonstrate that the proposed method consistently outperforms single-view and classical baselines, achieving approximately 40% mean accuracy in 7-class subject-independent emotion recognition. The code has been released at https://github.com/Danielz-z/LGF-EEG-Emotion.",
"arxiv_id": "2601.08094",
"authors": [
"Zheng Zhou",
"Isabella McEvoy",
"Camilo E. Valderrama"
],
"categories": [
"cs.LG",
"cs.AI"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"title": "Local-Global Feature Fusion for Subject-Independent EEG Emotion Recognition",
"url": "https://arxiv.org/abs/2601.08094",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "8fe74292-1aec-48c2-8be8-454d4663a74a",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}