Research map / Image generation / Super-resolution
Scene Prior Filtering for Depth Super-Resolution
Zhengxue Wang, Zhiqiang Yan, Ming–Hsuan Yang, Jinshan Pan, Guangwei Gao, Ying Yu Tai, Jian Yang
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 normal and semantic prior branches (ablation) | SGNet · NYU-v2 x16 | RMSE: 4.77 -> 4.70 | Improved |
| add normal and semantic prior branches (ablation) | SUFT · Lu x16 | RMSE: 3.92 -> 4.07 | Hurt |
| add normal and semantic prior branches (ablation) | FDSR · Lu x16 | RMSE: 5.00 -> 4.59 | Improved |
| add normal and semantic prior branches (ablation) | SPFNet · NYU-v2 x16 | RMSE: 4.73 -> 4.55 | Improved |
| add normal prior and semantic prior to RGB-only baseline (ablation) | SPFNet · Lu x16 | RMSE: 0.25cm | Improved |
| add depth-to-prior filtering (ablation) | MGF baseline · NYU-v2 x8 | RMSE: 2.59 -> 2.51 | Improved |
| add prior-to-depth filtering (ablation) | MGF baseline · NYU-v2 x8 | RMSE: 2.59 -> 2.48 | Improved |
| combine P2D then D2P instead of P2D only (ablation) | MGF · NYU-v2 x8 | RMSE: 2.48 -> 2.40 | Improved |
| add similarity weights to guide kernel generation (ablation) | MGF · NYU-v2 x8 | RMSE: 2.40 -> 2.36 | Improved |
| add APP and OPE modules (ablation) | SPFNet baseline · NYU-v2 | RMSE: 0.3cm | Improved |
| add similarity computation in APP and OPE (ablation) | SPFNet baseline · NYU-v2 | RMSE: 0.22cm | Improved |
| use SAM instead of MobileSAM (ablation) | SPFNet · Lu x16 | RMSE: 3.30 -> 3.25 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DORNet — DORNet: A Degradation Oriented and Regularized Network for Blind Depth Super-Resolution (CVPR 2025) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Recurrent Structure Attention Guidance for Depth Super-resolution (AAAI 2023 · similarity 0.89)
- Intrinsic Phase-Preserving Networks for Depth Super Resolution (AAAI 2024 · similarity 0.88)
- Learning Scene Structure Guidance via Cross-Task Knowledge Transfer for Single Depth Super-Resolution (CVPR 2021 · similarity 0.88)
- SGNet: Structure Guided Network via Gradient-Frequency Awareness for Depth Map Super-resolution (AAAI 2024 · similarity 0.87)
- The Power of Context: How Multimodality Improves Image Super-Resolution (CVPR 2025 · similarity 0.87)
- PM-INR: Prior-Rich Multi-Modal Implicit Large-Scale Scene Neural Representation (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.
