Fusion of Physiological and Behavioural Signals on SPD Manifolds with Application to Stress and Pain Detection. Wu, Y., Daoudi, M., Amad, A., Sparrow, L., & D'Hondt, F. Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, 2022-October:2949-2955, Institute of Electrical and Electronics Engineers Inc., 2022.
Fusion of Physiological and Behavioural Signals on SPD Manifolds with Application to Stress and Pain Detection [link]Paper  doi  abstract   bibtex   1 download  
Existing multimodal stress/pain recognition approaches generally extract features from different modalities independently and thus ignore cross-modality correlations. This paper proposes a novel geometric framework for multimodal stress/pain detection utilizing Symmetric Positive Definite (SPD) matrices as a representation that incorporates the correlation relationship of physiological and behavioural signals from covariance and cross-covariance. Considering the non-linearity of the Riemannian manifold of SPD matrices, well-known machine learning techniques are not suited to classify these matrices. Therefore, a tangent space mapping method is adopted to map the derived SPD matrix sequences to the vector sequences in the tangent space where the LSTM-based network can be applied for classification. The proposed framework has been evaluated on two public multimodal datasets, achieving both the state-of-the-art results for stress and pain detection tasks.
@article{Wu2022,
   abstract = {Existing multimodal stress/pain recognition approaches generally extract features from different modalities independently and thus ignore cross-modality correlations. This paper proposes a novel geometric framework for multimodal stress/pain detection utilizing Symmetric Positive Definite (SPD) matrices as a representation that incorporates the correlation relationship of physiological and behavioural signals from covariance and cross-covariance. Considering the non-linearity of the Riemannian manifold of SPD matrices, well-known machine learning techniques are not suited to classify these matrices. Therefore, a tangent space mapping method is adopted to map the derived SPD matrix sequences to the vector sequences in the tangent space where the LSTM-based network can be applied for classification. The proposed framework has been evaluated on two public multimodal datasets, achieving both the state-of-the-art results for stress and pain detection tasks.},
   author = {Yujin Wu and Mohamed Daoudi and Ali Amad and Laurent Sparrow and Fabien D'Hondt},
   doi = {10.1109/SMC53654.2022.9945260},
   isbn = {9781665452588},
   issn = {1062922X},
   journal = {Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics},
   keywords = {covariance matrix,multimodal fusion,pain detection,stress detection,symmetric positive definite manifold.},
   pages = {2949-2955},
   publisher = {Institute of Electrical and Electronics Engineers Inc.},
   title = {Fusion of Physiological and Behavioural Signals on SPD Manifolds with Application to Stress and Pain Detection},
   volume = {2022-October},
   url = {https://ieeexplore.ieee.org/document/9945260},
   year = {2022}
}

Downloads: 1