Research map / Learning theory and algorithms / Matrix and tensor methods
Dimensionality Reduction on SPD Manifolds: The Emergence of Geometry-Aware Methods
Mehrtash Tafazzoli Harandi, Mathieu Salzmann, Richard I. Hartley
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 |
|---|---|---|---|
| apply supervised SPD dimensionality reduction before kernel sparse coding with Stein divergence (ablation) | kSC-S · UIUC materials | accuracy (%): 52.8 -> 66.6 | Improved |
| apply supervised SPD dimensionality reduction before nearest neighbour with AIRM (ablation) | NN-AIRM · UIUC materials | accuracy (%): 35.6 -> 58.3 | Improved |
| apply supervised SPD dimensionality reduction before kernel sparse coding with log-Euclidean (ablation) | kSC-lE · UIUC materials | accuracy (%): 57.7 -> 63.9 | Improved |
| apply supervised SPD dimensionality reduction before kernel sparse coding with Stein divergence (ablation) | kSC-S · HDM05-MOCAP | accuracy (%): 76.1 -> 81.9 | Improved |
| apply supervised SPD dimensionality reduction before nearest neighbour with Stein divergence (ablation) | NN-S · HDM05-MOCAP | accuracy (%): 61.7 -> 68.6 | Improved |
| apply supervised SPD dimensionality reduction before nearest neighbour with Stein divergence (ablation) | NN-S · YTC face recognition | accuracy (%): 45.4 -> 72.8 | Improved |
| apply supervised SPD dimensionality reduction before nearest neighbour with log-Euclidean (ablation) | NN-lE · YTC face recognition | accuracy (%): 39.8 -> 55.0 | Improved |
| apply supervised SPD dimensionality reduction before kernel sparse coding with Stein divergence (ablation) | kSC-S · YTC face recognition | accuracy (%): 78.0 -> 80.1 | Improved |
| apply unsupervised SPD dimensionality reduction before kernel k-means with log-Euclidean (ablation) | kKM-lE · Keck gesture clustering | clustering accuracy (%): 71.3 -> 83.2 | Improved |
| apply unsupervised SPD dimensionality reduction before k-means with log-Euclidean (ablation) | KM-lE · Keck gesture clustering | clustering accuracy (%): 62.7 -> 75.3 | Improved |
| apply unsupervised SPD dimensionality reduction before k-means with Jeffrey divergence (ablation) | KM-J · Keck gesture clustering | clustering accuracy (%): 53.8 -> 55.3 | Improved |
| use Stein divergence or AIRM nearest neighbour on high-dimensional manifold instead of kernel sparse coding | NN-AIRM / NN-S · UIUC materials | accuracy (%): 52.8 -> 35.8 | Hurt |
Closest papers by meaning
- A Riemannian Network for SPD Matrix Learning (AAAI 2017 · similarity 0.90)
- Riemannian Local Mechanism for SPD Neural Networks (AAAI 2023 · similarity 0.89)
- Learning to Optimize on SPD Manifolds (CVPR 2020 · similarity 0.89)
- Deep Manifold Learning of Symmetric Positive Definite Matrices with Application to Face Recognition (AAAI 2017 · similarity 0.89)
- Approximate Log-Hilbert-Schmidt Distances Between Covariance Operators for Image Classification (CVPR 2016 · similarity 0.89)
- Matrix Manifold Neural Networks++ (ICLR 2024 · similarity 0.88)
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.
