Research map / Learning theory and algorithms / Matrix and tensor methods
Guaranteed Tensor Recovery Fused Low-rankness and Smoothness
Hailin Wang, Jiangjun Peng, Wenjin Qin, Jianjun Wang, Deyu Meng
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 the joint L+S t-CTV regularizer instead of TNN+TV (TNN-TV) | tensor completion · Color image inpainting, SR 3% | PSNR: 17.66 -> 21.32 | Improved |
| use t-CTV instead of TNN-TV at moderate sampling | tensor completion · Color image inpainting, SR 20% | PSNR: 24.71 -> 27.18 | Improved |
| use t-CTV instead of KBR at high sampling | tensor completion · Color image inpainting, SR 80% | PSNR: 40.16 -> 40.55 | Improved |
| use t-CTV instead of pure low-rank TNN | tensor completion · Color image inpainting, SR 10% | PSNR: 19.16 -> 24.59 | Improved |
| use t-CTV instead of TNN-TV | tensor completion · HSI inpainting SR 5% | ERGAS (lower better): 269.6 -> 70.15 | Improved |
| use t-CTV instead of KBR | tensor completion · HSI inpainting SR 5% | ERGAS (lower better): 105.9 -> 70.15 | Improved |
| use t-CTV instead of TNN+TV and TNN | TRPCA · Synthetic phase transition | correct recovery area: 32.87% | Improved |
| add TV term to TNN for TRPCA (TNN+TV vs TNN) | TNN TRPCA · Synthetic phase transition | correct recovery area | Hurt |
| use t-CTV (single regularizer) instead of TNN+TV with tuned trade-off | tensor completion · Synthetic phase transition | recovery success rate | Improved |
| use t-CTV in extreme missing rate | tensor completion · Color image inpainting, 99.5% missing | visual quality | Improved |
| run t-CTV vs TNN in compute time | tensor completion · Color image inpainting, SR 3% | Time/s: 5.96 -> 17.13 | Hurt |
| change transform in t-SVD among DFT, DCT, random orthogonal (ablation) | t-CTV-TC · Synthetic exact recovery | RelErr: 4.54e-7, 6.01e-6, 4.73e-7 | No clear effect |
Closest papers by meaning
- Tensor Compressive Sensing Fused Low-Rankness and Local-Smoothness (AAAI 2023 · similarity 0.91)
- Robust Low-Rank Tensor Recovery with Rectification and Alignment (TPAMI 2019 · similarity 0.90)
- Handling The Non-Smooth Challenge in Tensor SVD: A Multi-Objective Tensor Recovery Framework (ECCV 2024 · similarity 0.90)
- Low-Rank Tensor Completion: A Pseudo-Bayesian Learning Approach (ICCV 2017 · similarity 0.89)
- Low-Rank Tensor Completion with Total Variation for Visual Data Inpainting (AAAI 2017 · similarity 0.89)
- Deep Rank-One Tensor Functional Factorization for Multi-Dimensional Data Recovery (AAAI 2025 · similarity 0.87)
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.
