Research map / Learning theory and algorithms / Matrix and tensor methods
Efficient Robust Principal Component Analysis via Block Krylov Iteration and CUR Decomposition
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use eRPCA instead of ADW-RPCA | RPCA for foreground-background separation · CAVIAR1 (one frame) | PSNR: 34.4853 -> 34.8626 | Helped |
| Use eRPCA instead of IRCUR | RPCA for foreground-background separation · CAVIAR1 (one frame) | TIME(s): 0.007564 -> 0.034198 | Hurt |
| Use eRPCA instead of IRCUR | RPCA for foreground-background separation · CAVIAR1 (one frame) | pEPs%: 0.2828 -> 0.2686 | Helped |
| Use eRPCA instead of IRCUR | RPCA for foreground-background separation · CAVIAR1 (one frame) | MSSSIM: 0.9917 -> 0.9917 | No clear effect |
| Replace GDE rank estimation with KGDE using block Krylov iteration | Rank estimation · Ipi ceria3d | runtime: 91.180 -> 0.234 | Helped |
| Replace GDE rank estimation with KGDE using block Krylov iteration | Rank estimation · Ipi ceria3d | estimated rank: 5 -> 3 | Mixed |
| Use eRPCA instead of other RPCA methods | Synthetic RPCA · Synthetic data with corruption rate 0.7 and large rank | reconstruction error | Hurt |
| Use eRPCA instead of other RPCA methods | Synthetic RPCA · Synthetic data across tested corruption rates | estimated rank | Helped |
| Use eRPCA instead of PSSV | RPCA for face-image shadow removal · yale05 | runtime: 0.973s -> 0.071s | Helped |
| Use eRPCA instead of PSSV | RPCA for face-image shadow removal · yaleB02 | runtime: 0.962s -> 0.081s | Helped |
Closest papers by meaning
- Adaptive Rank Estimate in Robust Principal Component Analysis (CVPR 2021 · similarity 0.92)
- RES-PCA: A Scalable Approach to Recovering Low-Rank Matrices (CVPR 2019 · similarity 0.90)
- A Nonconvex Projection Method for Robust PCA (AAAI 2019 · similarity 0.87)
- Fast and Provable Nonconvex Tensor RPCA (ICML 2022 · similarity 0.86)
- Efficient Sparse PCA via Block-Diagonalization (ICLR 2025 · similarity 0.85)
- Fun with Flags: Robust Principal Directions via Flag Manifolds (CVPR 2024 · 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.