Research map / Learning theory and algorithms / Matrix and tensor methods
Partial Sum Minimization of Singular Values in Robust PCA: Algorithm and Applications
Tae-Hyun Oh, Yu‐Wing Tai, Jean‐Charles Bazin, Hyeongwoo Kim, In So Kweon
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 |
|---|---|---|---|
| replace nuclear norm RPCA (Wu et al.) with partial sum of singular values | RPCA photometric stereo · Bunny, 5 images, 5% corruption | mean normal error (degrees): 27.88 -> 7.06 | Improved |
| replace nuclear norm RPCA (Wu et al.) with partial sum of singular values | RPCA photometric stereo · Bunny, 12 images, 5% corruption | mean normal error (degrees): 9.95 -> 5.09 | Improved |
| replace least squares with PSSV (no light calibration) | photometric stereo · Bunny, 8 images | mean normal error (degrees): 9.03 -> 5.87 | Improved |
| replace least squares with PSSV (no light calibration) | photometric stereo · Bunny, 12 images | max normal error (degrees): 130.04 -> 76.86 | Improved |
| replace truncated nuclear norm local approximation (Hu et al.) with direct PSSV minimization | matrix completion · Image recovery | iterations: 755 -> 183 | Improved |
| replace truncated nuclear norm local approximation (Hu et al.) with direct PSSV minimization | matrix completion · Image recovery | time (sec): 18.60 -> 4.06 | Improved |
| replace truncated nuclear norm (Hu et al.) with PSSV | matrix completion · Image recovery | PSNR: 30.428 -> 30.429 | No clear effect |
| increase number of input images from 5 to 12 | Wu et al. RPCA · Bunny photometric stereo | mean normal error (degrees): 27.88 -> 9.95 | Improved |
| add partial-sum objective with soft target rank, versus hard rank matrix factorization | PSSV · synthetic low-rank with outliers | NRMSE | Improved |
Closest papers by meaning
- Adaptive Rank Estimate in Robust Principal Component Analysis (CVPR 2021 · similarity 0.91)
- Projected Robust PCA with Application to Smooth Image Recovery (JMLR 2022 · similarity 0.90)
- RES-PCA: A Scalable Approach to Recovering Low-Rank Matrices (CVPR 2019 · similarity 0.90)
- Bilinear Factor Matrix Norm Minimization for Robust PCA: Algorithms and Applications (TPAMI 2017 · similarity 0.90)
- Side Information in Robust Principal Component Analysis: Algorithms and Applications (ICCV 2017 · similarity 0.89)
- Tensor Robust Principal Component Analysis: Exact Recovery of Corrupted Low-Rank Tensors via Convex Optimization (CVPR 2016 · similarity 0.89)
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.
