Research map / Label-efficient and robust learning / Recognition and long-tail
A Novel Space-Time Representation on the Positive Semidefinite Cone for Facial Expression Recognition
Anis Kacem, Mohamed Daoudi, Boulbaba Ben Amor, Juan Carlos Alvarez-Paiva
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add covariance (S+(2,n) instead of Grassmann G(2,n) only) (ablation) | Trajectory on S+(2,n) with ppfSVM · CK+ | accuracy: 94.25 -> 96.87 | Improved |
| add covariance (S+(2,n) instead of Grassmann G(2,n) only) (ablation) | Trajectory on S+(2,n) with ppfSVM · MMI | accuracy: 78.18 -> 79.19 | Improved |
| add covariance (S+(2,n) instead of Grassmann G(2,n) only) (ablation) | Trajectory on S+(2,n) with ppfSVM · Oulu-CASIA | accuracy: 80.0 -> 83.13 | Improved |
| add covariance (S+(2,n) instead of Grassmann G(2,n) only) (ablation) | Trajectory on S+(2,n) with ppfSVM · AFEW (validation, EmotiW'13) | accuracy: 39.1 -> 39.94 | Improved |
| use closeness d_S+ instead of flat Frobenius distance (ablation) | Trajectory on S+(2,n) with ppfSVM · CK+ | accuracy: 93.78 -> 96.87 | Improved |
| use closeness d_S+ instead of distance in P_n (ablation) | Trajectory on S+(2,n) with ppfSVM · CK+ | accuracy: 92.92 -> 96.87 | Improved |
| add DTW temporal alignment (ablation) | Trajectory on S+(2,n) with ppfSVM · CK+ | accuracy: 90.94 -> 96.87 | Improved |
| add DTW temporal alignment (ablation) | Trajectory on S+(2,n) with ppfSVM · MMI | accuracy: 66.93 -> 79.19 | Improved |
| add adaptive re-sampling (ablation) | Trajectory on S+(2,n) with ppfSVM · MMI | accuracy: 74.72 -> 79.19 | Improved |
| add adaptive re-sampling (ablation) | Trajectory on S+(2,n) with ppfSVM · AFEW (validation, EmotiW'13) | accuracy: 36.81 -> 39.94 | Improved |
| use ppfSVM instead of K-NN (ablation) | Trajectory on S+(2,n) with ppfSVM · CK+ | accuracy: 88.97 -> 96.87 | Improved |
| use ppfSVM instead of K-NN (ablation) | Trajectory on S+(2,n) with ppfSVM · AFEW (validation, EmotiW'13) | accuracy: 29.77 -> 39.94 | Improved |
Closest papers by meaning
- A Novel Geometric Framework on Gram Matrix Trajectories for Human Behavior Understanding (TPAMI 2018 · similarity 0.93)
- Sparse Coding of Shape Trajectories for Facial Expression and Action Recognition (TPAMI 2019 · similarity 0.89)
- Cross Euclidean-to-Riemannian Metric Learning with Application to Face Recognition from Video (TPAMI 2017 · similarity 0.87)
- Multilinear Modelling of Faces and Expressions (TPAMI 2020 · similarity 0.86)
- TEASER: Token Enhanced Spatial Modeling for Expressions Reconstruction (ICLR 2025 · similarity 0.85)
- Lifting Scheme-Based Implicit Disentanglement of Emotion-Related Facial Dynamics in the Wild (AAAI 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
