Research map / 3D vision / Depth and stereo
RoutedFusion: Learning Real-Time Depth Map Fusion
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 |
|---|---|---|---|
| Replace linear TSDF averaging with learned nonlinear TSDF updates | RoutedFusion · ShapeNet test | MAD: 0.0078 -> 0.0050 | Helped |
| Replace linear TSDF averaging with learned nonlinear TSDF updates | RoutedFusion · ShapeNet test | Accuracy [%]: 88.06 -> 94.77 | Helped |
| Replace linear TSDF averaging with learned nonlinear TSDF updates | RoutedFusion · ShapeNet test | IoU: 0.659 -> 0.785 | Helped |
| Add the depth routing network (ablation) | RoutedFusion · ShapeNet test | MAD: 0.0051 -> 0.0050 | Helped |
| Add the depth routing network (ablation) | RoutedFusion · ShapeNet test | Accuracy [%]: 93.91 -> 94.77 | Helped |
| Add the depth routing network (ablation) | RoutedFusion · ShapeNet test | IoU: 0.765 -> 0.785 | Helped |
| Add the depth routing network (ablation) | RoutedFusion · ShapeNet test | MSE [e-05]: 5.9 -> 5.9 | No clear effect |
| Preprocess standard TSDF fusion with the depth routing network | TSDF Fusion · ShapeNet test | MAD: 0.0078 -> 0.0084 | Hurt |
| Add the depth routing network (ablation) | RoutedFusion · 3D Scene Data — Lounge | MAD [mm]: 0.0055 -> 0.0051 | Helped |
| Add the depth routing network (ablation) | RoutedFusion · 3D Scene Data — Copyroom | MAD [mm]: 0.0057 -> 0.0051 | Helped |
| Add the depth routing network (ablation) | RoutedFusion · 3D Scene Data — Stonewall | MAD [mm]: 0.0047 -> 0.0043 | Helped |
| Add the depth routing network (ablation) | RoutedFusion · 3D Scene Data — Cactusgarden | MAD [mm]: 0.0055 -> 0.0052 | Helped |
Closest papers by meaning
- NeuralFusion: Online Depth Fusion in Latent Space (CVPR 2021 · similarity 0.91)
- Learning Non-Volumetric Depth Fusion Using Successive Reprojections (CVPR 2019 · similarity 0.89)
- Self-Distilled Depth Refinement with Noisy Poisson Fusion (NeurIPS 2024 · similarity 0.87)
- StereoDRNet: Dilated Residual StereoNet (CVPR 2019 · similarity 0.87)
- Improving Depth Completion via Depth Feature Upsampling (CVPR 2024 · similarity 0.86)
- PatchFusion: An End-to-End Tile-Based Framework for High-Resolution Monocular Metric Depth Estimation (CVPR 2024 · similarity 0.86)
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.