Research map / Image generation / Super-resolution
Structure Flow-Guided Network for Real Depth Super-resolution
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 the full SFG framework rather than the strongest listed baseline | Real-world DSR · RGB-D-D hand-filled | RMSE: 5.08 -> 3.88 | Helped |
| Use the full SFG framework rather than the strongest listed baseline | Real-world DSR · RGB-D-D incomplete | RMSE: 5.43 -> 4.79 | Helped |
| Use the full SFG framework rather than the strongest listed baseline | Real-world DSR · RGB-D-D noisy | RMSE: 5.16 -> 4.45 | Helped |
| Use the full SFG framework rather than the strongest listed baseline | Real-world DSR · ToFMark | RMSE: 0.26 -> 0.25 | Helped |
| Use the full SFG framework rather than the strongest listed baseline | Synthetic DSR · NYU-v2 ×4 | RMSE: 1.49 -> 1.45 | Helped |
| Use the full SFG framework rather than the strongest listed baseline | Synthetic DSR · NYU-v2 ×8 | RMSE: 2.73 -> 2.84 | Hurt |
| Use the full SFG framework rather than the strongest listed baseline | Synthetic DSR · NYU-v2 ×16 | RMSE: 5.11 -> 5.56 | Hurt |
| Add trilateral self-attention to CFUNet (ablation) | CFUNet on RGB-D-D · RGB-D-D | RMSE: 4.34 -> 4.22 | Helped |
| Add cross-attention to CFUNet (ablation) | CFUNet on RGB-D-D · RGB-D-D | RMSE: 4.57 -> 4.22 | Helped |
| Add PEANet for edge refinement (ablation) | SFG · RGB-D-D | RMSE: 4.22 -> 3.88 | Helped |
| Add PEANet for edge refinement (ablation) | SFG · NYU-v2 ×4 | RMSE: 1.82 -> 1.45 | Helped |
| Add PEANet for edge refinement (ablation) | SFG · NYU-v2 ×8 | RMSE: 3.76 -> 2.84 | Helped |
Closest papers by meaning
- SGNet: Structure Guided Network via Gradient-Frequency Awareness for Depth Map Super-resolution (AAAI 2024 · similarity 0.91)
- Intrinsic Phase-Preserving Networks for Depth Super Resolution (AAAI 2024 · similarity 0.89)
- DegBins: Degradation-Driven Binning for Depth Super-Resolution (NeurIPS 2026 · similarity 0.88)
- DORNet: A Degradation Oriented and Regularized Network for Blind Depth Super-Resolution (CVPR 2025 · similarity 0.87)
- Improving Depth Completion via Depth Feature Upsampling (CVPR 2024 · similarity 0.86)
- Noise-Started One-Step Image Super-Resolution via LR-Conditioned SplitMeanFlow and GAN Refinement (NeurIPS 2026 · similarity 0.84)
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.