Research map / 3D vision / Depth and stereo
Discontinuity-aware Normal Integration for Generic Central Camera Models
Francesco Milano, Manuel López-Antequera, Naina Dhingra, Roland Siegwart, Robert Thiel
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 iterative discontinuity (alpha) computation (ablation) | Ours · DiLiGenT harvest | MADE (mm): 4.83 -> 0.73 | Improved |
| add iterative discontinuity (alpha) computation (ablation) | Ours · DiLiGenT goblet | MADE (mm): 6.56 -> 4.72 | Improved |
| add iterative discontinuity (alpha) computation (ablation) | Ours · DiLiGenT reading | MADE (mm): 0.12 -> 0.17 | Hurt |
| add iterative discontinuity (alpha) computation (ablation) | Ours · DiLiGenT bear | MADE (mm): 0.07 -> 0.03 | Improved |
| run more iterations (150 to 1200) | Ours · DiLiGenT harvest | MADE (mm): 1.35 -> 0.73 | Improved |
| use ray-direction local planarity formulation instead of BiNI | Ours · DiLiGenT harvest | MADE (mm): 1.84 -> 0.73 | Improved |
| use ray-direction local planarity formulation instead of BiNI | Ours · DiLiGenT bear residual | absolute residual on ground-truth log-depth: 3.72 -> 0.82 | Improved |
| provide ground-truth discontinuity values | Ours · DiLiGenT harvest | MADE (mm): 0.73 -> 0.34 | Improved |
| include the n^T tau term in the cost function weighting (ablation) | Ours w/o alpha · DiLiGenT pot1 | MADE (mm): 14.40 -> 0.49 | Improved |
| include the n^T tau term in the cost function weighting (ablation) | Ours w/o alpha · DiLiGenT bear | MADE (mm): 0.48 -> 0.07 | Improved |
| lower the f multiplier in the bilateral weights from 3000 to 1000 (ablation) | Ours w/o alpha · DiLiGenT buddha | MADE (mm): 0.27 -> 1.41 | Hurt |
| lower the f multiplier in the bilateral weights from 3000 to 1000 (ablation) | Ours w/o alpha · DiLiGenT harvest | MADE (mm): 3.89 -> 2.51 | Mixed |
Closest papers by meaning
- Normal Integration via Inverse Plane Fitting With Minimum Point-to-Plane Distance (CVPR 2021 · similarity 0.91)
- An Adaptive Screen-Space Meshing Approach for Normal Integration (ECCV 2024 · similarity 0.90)
- Feature-Preserving Mesh Decimation for Normal Integration (CVPR 2025 · similarity 0.88)
- Discontinuity-preserving Normal Integration with Auxiliary Edges (CVPR 2024 · similarity 0.87)
- PMNI: Pose-free Multi-view Normal Integration for Reflective and Textureless Surface Reconstruction (CVPR 2025 · similarity 0.86)
- RNb-NeuS: Reflectance and Normal-based Multi-View 3D Reconstruction (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.
