Research map / Label-efficient and robust learning / Recognition and long-tail
Nuclear Norm Based Matrix Regression with Applications to Face Recognition with Occlusion and Illumination Changes
Jian Jun Yang, Lei Luo, Jianjun Qian, Ying Yu Tai, Fanlong Zhang, Yong Dong Xu
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 |
|---|---|---|---|
| use nuclear norm matrix regression instead of SRC for scarf disguise | NMR · AR (scarf) | recognition rate (%): 57.6 -> 73.5 | Improved |
| use NMR instead of SRC on sunglasses (low occlusion) | NMR · AR (sunglasses) | recognition rate (%): 94.4 -> 96.9 | No clear effect |
| use NMR instead of RSC at 60% random block occlusion | NMR · Extended Yale B (random block occlusion 60%) | recognition rate (%): 86.4%, 22.8% higher than RSC | Improved |
| use NMR instead of SRC/RSC/HQ_M at low occlusion levels (30% or less) | NMR · Extended Yale B (baboon block) | recognition rate | No clear effect |
| use NMR instead of SRC/RSC/HQ_M/SSEC at 50% or more occlusion | NMR · Extended Yale B (baboon block) | recognition rate | Improved |
| use NMR instead of RSC under Multi-PIE illumination | NMR · Multi-PIE (Session 3) | recognition rate (%): 75.3 -> 77.9 | Improved |
| use NMR instead of CRC on FRGC | NMR · FRGC (220 persons) | recognition rate (%): 92.2 -> 93.3 | Improved |
| use occlusion-specific SSEC instead of NMR for illumination changes | SSEC · Multi-PIE (Session 4) | recognition rate (%): 84.5 -> 59.1 | Hurt |
| vary NMR regression parameter within 0.5 to 1 (ablation) | NMR · Extended Yale B, AR, Multi-PIE, FRGC | recognition rate | No clear effect |
Closest papers by meaning
- Enhanced Group Sparse Regularized Nonconvex Regression for Face Recognition (TPAMI 2020 · similarity 0.85)
- Low-Rank Laplacian-Uniform Mixed Model for Robust Face Recognition (CVPR 2019 · similarity 0.85)
- Bilinear Factor Matrix Norm Minimization for Robust PCA: Algorithms and Applications (TPAMI 2017 · similarity 0.84)
- Two-Dimensional PCA with F-Norm Minimization (AAAI 2017 · similarity 0.83)
- Two-Dimensional Whitening Reconstruction for Enhancing Robustness of Principal Component Analysis (TPAMI 2016 · similarity 0.83)
- SDD-FIQA: Unsupervised Face Image Quality Assessment With Similarity Distribution Distance (CVPR 2021 · similarity 0.82)
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.
