Research map / Image generation / Image restoration
MARLow: A Joint Multiplanar Autoregressive and Low-Rank Approach for Image Completion
Mading Li, Jiaying Liu, Zhiwei Xiong, Xiaoyan Sun, Zongming Guo
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 |
|---|---|---|---|
| add the multiplanar AR model to low-rank minimization (ablation) | MARLow · Example image, low-rank only vs MARLow | PSNR: 19.48 dB -> 19.95 dB | Improved |
| add the multiplanar AR model to low-rank minimization (ablation) | MARLow · Example image, low-rank only vs MARLow | SSIM: 0.8933 -> 0.8980 | Improved |
| process color channels simultaneously instead of separately (ablation) | MARLow · Color completion, 80% missing | PSNR | Improved |
| replace JSM with MARLow | JSM · Gray-scale completion, 80% missing average | PSNR: 30.35 -> 31.41 | Improved |
| replace BPFA with MARLow | BPFA · Color completion average | PSNR: 26.98 -> 29.76 | Improved |
| replace BPFA with MARLow at 90% missing | BPFA · Color image Woman, 90% missing | PSNR: 24.20 -> 30.12 | Improved |
| replace JSM with MARLow | JSM · Text removal average | PSNR: 35.30 -> 36.89 | Improved |
| replace DCC with MARLow | DCC · Image interpolation average | PSNR: 30.13 -> 30.16 | No clear effect |
| replace DCC with MARLow | DCC · Image interpolation average | SSIM: 0.9009 -> 0.8998 | Hurt |
| replace DCC with MARLow on Cameraman interpolation | DCC · Cameraman interpolation | PSNR: 25.92 -> 25.36 | Hurt |
Closest papers by meaning
- Multi-Matrix Completion: A Novel Framework for Structurally Missing Elements (TPAMI 2025 · similarity 0.85)
- Pluralistic Image Completion with Gaussian Mixture Models (NeurIPS 2022 · similarity 0.85)
- From Missing Pieces to Masterpieces: Image Completion With Context-Adaptive Diffusion (TPAMI 2025 · similarity 0.85)
- NFIG: Multi-Scale Autoregressive Image Generation via Frequency Ordering (NeurIPS 2025 · similarity 0.84)
- Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction (NeurIPS 2024 · similarity 0.83)
- MaIR: A Locality- and Continuity-Preserving Mamba for Image Restoration (CVPR 2025 · 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.
