Research map / 3D vision / Face and shape
MNSRNet: Multimodal Transformer Network for 3D Surface Super-Resolution
Check 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 MNSRNet instead of 3DASR | 3D surface super-resolution · Hybrid testing datasets, ×2 | PSNR: 28.3017 -> 28.7662 | Helped |
| Use MNSRNet instead of IPT | 3D surface super-resolution · Hybrid testing datasets, ×4 | PSNR: 23.6268 -> 23.7961 | Helped |
| Add MPS, cmTA, and cmAF rather than omitting all three modules (ablation) | MNSRNet · DiLiGenT, ×4 | PSNR: 22.3302 -> 23.7961 | Helped |
| Add cmTA while retaining MPS and cmAF (ablation) | MNSRNet · DiLiGenT, ×4 | PSNR: 23.6824 -> 23.7961 | Helped |
| Add cmAF while retaining MPS and cmTA (ablation) | MNSRNet · DiLiGenT, ×4 | PSNR: 23.4994 -> 23.7961 | Helped |
| Add the texture modality to normal input (ablation) | MNSRNet · ×2 | PSNR: 27.9376 -> 28.3151 | Helped |
| Add the depth modality to normal input (ablation) | MNSRNet · ×2 | PSNR: 27.9376 -> 28.3816 | Helped |
| Use texture and depth together with normal instead of normal alone (ablation) | MNSRNet · ×2 | PSNR: 27.9376 -> 28.7662 | Helped |
Closest papers by meaning
- SuperNormal: Neural Surface Reconstruction via Multi-View Normal Integration (CVPR 2024 · similarity 0.89)
- 3D Appearance Super-Resolution With Deep Learning (CVPR 2019 · similarity 0.88)
- Just Look at the Image: Viewpoint-Specific Surface Normal Prediction for Improved Multi-View Reconstruction (CVPR 2016 · similarity 0.86)
- GenS: Generalizable Neural Surface Reconstruction from Multi-View Images (NeurIPS 2023 · similarity 0.86)
- Multiview Neural Surface Reconstruction by Disentangling Geometry and Appearance (NeurIPS 2020 · similarity 0.86)
- PMNI: Pose-free Multi-view Normal Integration for Reflective and Textureless Surface Reconstruction (CVPR 2025 · 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.