Academic Journal
Automatic Estimation of Self-Reported Pain by Trajectory Analysis in the Manifold of Fixed Rank Positive Semi-Definite Matrices
| العنوان: | Automatic Estimation of Self-Reported Pain by Trajectory Analysis in the Manifold of Fixed Rank Positive Semi-Definite Matrices |
|---|---|
| المؤلفون: | Szczapa, Benjamin, Daoudi, Mohamed, Berretti, Stefano, Pala, Pietro, Bimbo, Alberto, Del, Hammal, Zakia |
| المساهمون: | Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 (CRIStAL), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS), Ecole nationale supérieure Mines-Télécom Lille Douai (IMT Nord Europe), Institut Mines-Télécom Paris (IMT), Università degli Studi di Firenze = University of Florence = Université de Florence (UniFI), Carnegie Mellon University Pittsburgh (CMU) |
| المصدر: | ISSN: 1949-3045 ; IEEE Transactions on Affective Computing ; https://hal.science/hal-03768792 ; IEEE Transactions on Affective Computing, 2022, 13 (4), pp.1813-1826. ⟨10.1109/TAFFC.2022.3207001⟩. |
| بيانات النشر: | HAL CCSD Institute of Electrical and Electronics Engineers |
| سنة النشر: | 2022 |
| المجموعة: | LillOA (HAL Lille Open Archive, Université de Lille) |
| مصطلحات موضوعية: | Pain estimation, Gram matrix, Facial landmarks, Fixed rank positive semi-definite matrices, Trajectory on a manifold, Learning on manifold, Shape analysis, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] |
| الوصف: | International audience ; We propose an automatic method to estimate self-reported pain based on facial landmarks extracted from videos. For each video sequence, we decompose the face into four different regions and the pain intensity is measured by modeling the dynamics of facial movement using the landmarks of these regions. A formulation based on Gram matrices is used for representing the trajectory of landmarks on the Riemannian manifold of symmetric positive semi-definite matrices of fixed rank. A curve fitting algorithm is used to smooth the trajectories and temporal alignment is performed to compute the similarity between the trajectories on the manifold. A Support Vector Regression classifier is then trained to encode extracted trajectories into pain intensity levels consistent with self-reported pain intensity measurement. Finally, a late fusion of the estimation for each region is performed to obtain the final predicted pain level. The proposed approach is evaluated on two publicly available datasets, the UNBCMcMaster Shoulder Pain Archive and the Biovid Heat Pain dataset. We compared our method to the state-of-the-art on both datasets using different testing protocols, showing the competitiveness of the proposed approach. |
| نوع الوثيقة: | article in journal/newspaper |
| اللغة: | English |
| Relation: | hal-03768792; https://hal.science/hal-03768792; https://hal.science/hal-03768792/document; https://hal.science/hal-03768792/file/TACMinorReview.pdf |
| DOI: | 10.1109/TAFFC.2022.3207001 |
| الاتاحة: | https://hal.science/hal-03768792 https://hal.science/hal-03768792/document https://hal.science/hal-03768792/file/TACMinorReview.pdf https://doi.org/10.1109/TAFFC.2022.3207001 |
| Rights: | info:eu-repo/semantics/OpenAccess |
| رقم الانضمام: | edsbas.F5926C7 |
| قاعدة البيانات: | BASE |
| DOI: | 10.1109/TAFFC.2022.3207001 |
|---|