Determining Levels of Affective States with Riemannian Geometry Applied to EEG Signals
Source abstract
Abstract
Emotion recognition from electroencephalography (EEG) often relies on Euclidean features that ignore the curved geometry of covariance matrices. We introduce a Riemannian-manifold pipeline which, combined with the Fisher Geodesic Minimum Distance to Mean (FgMDM) classifier, leverages the full geometry of symmetric positive-definite (SPD) EEG covariances. The approach applies an additional geodesic-mean contraction that stabilizes trial covariances before tangent space projection. Experiments on the five-class SEED-V dataset show high accuracy, robustness to session-to-session variability and improved interpretability relative to baselines. These results highlight Riemannian geometry as a powerful framework for emotion recognition with high-dimensional, non-stationary EEG.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Technology and Human Factors in Education and Health · Functional Brain Connectivity Studies
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Institute of Information Technology, Lodz University of Technology, Wólczańska 215, 90-924 Łódź, Poland
References (36)
Show source-provided references
1. Tibermacine, A., Tibermacine, I.E., Zouai, M., and Rabehi, A. (2024, January 14–15). EEG Classification Using Contrastive Learning and Riemannian Tangent Space Representations. Proceedings of the 2024 International Conference on Telecommunications and Intelligent Systems (ICTIS), Djelfa, Algeria.10.1109/ICTIS62692.2024.10894645 ↗
2. A Bi-Hemisphere Domain Adversarial Neural Network Model for EEG Emotion Recognition10.1109/TAFFC.2018.2885474 ↗
3. Li, T.H., Liu, W., Zheng, W.L., and Lu, B.L. (2019, January 20–23). Classification of Five Emotions from EEG and Eye Movement Signals: Discrimination Ability and Stability over Time. Proceedings of the 2019 9th International IEEE/EMBS Conference on Neural Engineering (NER), San Francisco, CA, USA.10.1109/NER.2019.8716943 ↗
4. Yan, X., Zhao, L.M., and Lu, B.L. (2021, January 20–24). Simplifying Multimodal Emotion Recognition with Single Eye Movement Modality. Proceedings of the 29th ACM International Conference on Multimedia (MM 21), Chengdu, China.10.1145/3474085.3475701 ↗
5. Classification of Covariance Matrices Using a Riemannian-Based Kernel for BCI Applications10.1016/j.neucom.2012.12.039 ↗
6. Abdel-Ghaffar, E.A., and Daoudi, M. (2020, January 6–8). Emotion Recognition from Multidimensional Electroencephalographic Signals on the Manifold of Symmetric Positive Definite Matrices. Proceedings of the 2020 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR), Shenzhen, China.10.1109/MIPR49039.2020.00078 ↗
7. Multiclass Brain–Computer Interface Classification by Riemannian Geometry10.1109/TBME.2011.2172210 ↗
8. Intrinsic Statistics on Riemannian Manifolds: Basic Tools for Geometric Measurements10.1007/s10851-006-6228-4 ↗
9. Riemannian Approaches in Brain-Computer Interfaces: A Review10.1109/TNSRE.2016.2627016 ↗
10. Improving BCI-Based Emotion Recognition by Combining EEG Feature Selection and Kernel Classifiers10.1016/j.eswa.2015.10.049 ↗
11. Liu, J., Wu, G., Luo, Y., Qiu, S., Yang, S., Li, W., and Bi, Y. (2018). EEG-Based Emotion Classification Using a Deep Neural Network and Sparse Autoencoder. Front. Neurosci., 12.
12. EEG-Based Emotion Recognition Using an End-to-End Regional-Asymmetric Convolutional Neural Network10.1016/j.knosys.2020.106243 ↗
13. Cui, F., Wang, R., Ding, W., Chen, Y., and Huang, L. (2020). A novel DE-CNN-BiLSTM multi-fusion model for EEG emotion recognition. Mathematics, 10.10.3390/math10040582 ↗
14. Hu, J. (2017). Automated detection of driver fatigue based on AdaBoost classifier with EEG signals. Front. Neurosci., 11.10.3389/fncom.2017.00072 ↗
15. Balconi, M., Acconito, C., and Angioletti, L. (2025). A Preliminary EEG Study on Persuasive Communication towards Groupness. Sci. Rep., 15.10.1038/s41598-025-90301-y ↗
16. Andreev, A., Cattan, G., and Congedo, M. (2025). The Riemannian Means Field Classifier for EEG-Based BCI Data. Sensors, 25.10.3390/s25072305 ↗
17. Manifold-Valued Models for Analysis of EEG Time Series Data10.1016/j.csda.2025.108168 ↗
18. Riemannian Geometry for EEG-Based Brain–Computer Interfaces: A Primer and a Review10.1080/2326263X.2017.1297192 ↗
19. Subject-Dependent Emotion Recognition System Based on Multidimensional Electroencephalographic Signals: A Riemannian Geometry Approach10.1109/ACCESS.2022.3147461 ↗
20. Al-Mashhadani, Z., Bayat, N., Kadhim, I.F., Choudhury, R., and Park, J.-H. (2023). The Efficacy and Utility of Lower-Dimensional Riemannian Geometry for EEG-Based Emotion Classification. Appl. Sci., 13.10.3390/app13148274 ↗
21. Zhang, D., Yao, L., Zhang, X., Wang, S., Chen, W., Boots, R., and Benatallah, B. (2020). Cascade and parallel convolutional recurrent neural networks on EEG-based intention recognition for brain computer interface. Proc. AAAI Conf. Artif., 1.
22. Gao, C., Uchitomi, H., and Miyake, Y. (2023). Cross-Sensory EEG Emotion Recognition with Filter Bank Riemannian Feature and Adversarial Domain Adaptation. Brain Sci., 13.10.3390/brainsci13091326 ↗
23. GREEN: A lightweight architecture using learnable wavelets and Riemannian geometry for biomarker exploration with EEG signals10.1016/j.patter.2025.101182 ↗
24. ERTNet: An Interpretable Transformer-Based Framework for EEG Emotion Recognition10.3389/fnins.2024.1320645 ↗
25. Lan, Y.-T., Liu, W., and Lu, B.-L. (2020, January 19–24). Multimodal Emotion Recognition Using Deep Generalized Canonical Correlation Analysis with an Attention Mechanism. Proceedings of the 2020 International Joint Conference on Neural Networks (IJCNN), Glasgow, UK.10.1109/IJCNN48605.2020.9207625 ↗
26. Individual Classification of Emotions Using EEG10.4236/jbise.2014.78061 ↗
27. The Discrete Complex Wavelet Approach to Phase Assignment and a New Test Bed for Related Methods10.1063/1.4906814 ↗
28. A Riemannian Geometry Approach to Reduced and Discriminative Covariance Estimation in Brain Computer Interfaces10.1109/TBME.2019.2912066 ↗
29. Generalized Learning Vector Quantization with Log–Euclidean Metric Learning on Symmetric Positive–Definite Manifold10.1109/TCYB.2022.3178412 ↗
30. Kobler, R.J., Hirayama, J.-I., Hehenberger, L., Lopes-Dias, C., Müller-Putz, G.R., and Kawanabe, M. (2021, January 1–5). On the Interpretation of Linear Riemannian Tangent Space Model Parameters in M/EEG. Proceedings of the 2021 43rd Annual International Conference of the IEEE Engineering in Medicine I & Biology Society (EMBC), Cancun, Mexico.10.1109/EMBC46164.2021.9630144 ↗
31. Yang, H., Wu, F., Zhang, N., Chen, C., Wei, A., and Peng, F. (2023, January 17–20). A Transfer Learning Method for Motor Imagery EEG Signals Classification Based on CCSP and Riemannian Tangent Space Mapping. Proceedings of the 2023 IEEE International Conference on Real-time Computing and Robotics (RCAR), Datong, China.10.1109/RCAR58764.2023.10249485 ↗
32. Liu, Y.K., Jiang, W.B., and Lu, B.L. (2022, January 18–23). Increasing the Stability of EEG-Based Emotion Recognition with a Variant of Neural Processes. Proceedings of the 2022 International Joint Conference on Neural Networks (IJCNN), Padua, Italy.10.1109/IJCNN55064.2022.9892825 ↗
33. Zhang, G., and Etemad, A. (2020). Spatio-Temporal EEG Representation Learning on Riemannian Manifold and Euclidean Space. arXiv, Available online: https://arxiv.org/abs/2008.08633.
34. Explainable Multi-Frequency and Multi-Region Fusion Model for Affective Brain–Computer Interfaces
35. Gupta, V., Kendre, T.P., Reddy, T.K., and Arora, V. (2022, January 24–27). Comparative Performance Analysis of Scalp EEG and Ear EEG based P300 Ambulatory Brain-Computer Interfaces using Riemannian Geometry and Convolutional Neural Networks. Proceedings of the 2022 National Conference on Communications (NCC), Mumbai, India.10.1109/NCC55593.2022.9806815 ↗
36. Ma, T.F., Zheng, W.L., and Lu, B.L. (2022, January 22–26). Few-shot Class-incremental Learning for EEG-based Emotion Recognition. Proceedings of the 29th International Conference on Neural Information Processing (ICONIP 2022), Virtual Event. Available online: https://bcmi.sjtu.edu.cn/home/blu/papers/2022/2022-15.pdf.