Research map / Image generation / Super-resolution
Recurrent Structure Attention Guidance for 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 |
|---|---|---|---|
| Replace hand-designed frequency decomposition with DCN (ablation) | RSAG baseline · Middlebury ×8 | MAD: 0.26 -> 0.24 | Helped |
| Replace hand-designed frequency decomposition with DCN (ablation) | RSAG baseline · NYU-v2 ×8 | RMSE: 3.60 -> 3.10 | Helped |
| Add HLF to baseline + DCN (ablation) | RSAG · Middlebury ×8 | MAD: 0.24 -> 0.23 | Helped |
| Add HLF to baseline + DCN (ablation) | RSAG · NYU-v2 ×8 | RMSE: 3.10 -> 3.02 | Helped |
| Add recurrent structure attention to baseline + DCN + HLF (ablation) | RSAG · Middlebury ×8 | MAD: 0.23 -> 0.19 | Helped |
| Add recurrent structure attention to baseline + DCN + HLF (ablation) | RSAG · NYU-v2 ×8 | RMSE: 3.02 -> 2.51 | Helped |
| Increase recurrent SA steps from 0 to 2 (ablation) | RSAG · Middlebury ×8 and NYU-v2 ×8 | MAD and RMSE | Helped |
| Increase recurrent SA steps beyond 2 (ablation) | RSAG · Middlebury ×8 and NYU-v2 ×8 | MAD and RMSE | Mixed |
| Use RSAG instead of CTKT | Depth super-resolution · NYU-v2 ×4 | RMSE: 1.49 -> 1.23 | Helped |
| Use RSAG instead of CTKT | Depth super-resolution · NYU-v2 ×8 | RMSE: 2.73 -> 2.51 | Helped |
| Use RSAG instead of CTKT | Depth super-resolution · NYU-v2 ×16 | RMSE: 5.11 -> 5.27 | Hurt |
| Use RSAG instead of FDSR | Depth super-resolution · Lu ×4, ×8, and ×16 | RMSE: 0.81 -> 0.79; 1.91 -> 1.67; 4.64 -> 4.30 | Helped |
Closest papers by meaning
- SGNet: Structure Guided Network via Gradient-Frequency Awareness for Depth Map Super-resolution (AAAI 2024 · similarity 0.91)
- Structure Flow-Guided Network for Real Depth Super-resolution (AAAI 2023 · similarity 0.90)
- Guided Depth Super-Resolution by Deep Anisotropic Diffusion (CVPR 2023 · similarity 0.88)
- Intrinsic Phase-Preserving Networks for Depth Super Resolution (AAAI 2024 · similarity 0.87)
- DegBins: Degradation-Driven Binning for Depth Super-Resolution (NeurIPS 2026 · similarity 0.85)
- ResDiff: Combining CNN and Diffusion Model for Image Super-resolution (AAAI 2024 · similarity 0.85)
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.