Research map / 3D vision / 3D generation
Semantically Informed Multiview Surface Refinement
Maros Blaha, Mathias Rothermel, Martin R. Oswald, Torsten Sattler, Audrey Richard, Jan D. Wegner, Marc Pollefeys, Konrad Schindler
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 semantic label consistency and class-specific shape priors to geometric mesh refinement | mesh refinement from semantic 3D reconstruction input · SynthCity3 A | mean distance to ground truth (m): 0.44 -> 0.38 | Improved |
| add semantic label consistency and class-specific shape priors to geometric mesh refinement | mesh refinement from semantic 3D reconstruction input · SynthCity3 B | mean distance to ground truth (m): 0.74 -> 0.62 | Improved |
| add semantic relabeling with shape priors to geometric refinement | mesh refinement from semantic 3D reconstruction input · SynthCity3 A | semantic average accuracy (%): 82.8 -> 88.8 | Improved |
| add semantic relabeling with shape priors to geometric refinement | mesh refinement from semantic 3D reconstruction input · SynthCity3 B | semantic average accuracy (%): 83.8 -> 90.0 | Improved |
| add semantic relabeling with shape priors to geometric refinement | mesh refinement from semantic 3D reconstruction input · Enschede A | semantic average accuracy (%): 78.8 -> 83.3 | Improved |
| add semantic relabeling with shape priors to geometric refinement | mesh refinement from semantic 3D reconstruction input · Enschede B | semantic overall accuracy (%): 90.6 -> 94.1 | Improved |
| add semantic relabeling with shape priors to geometric refinement | mesh refinement from semantic 3D reconstruction input · Dortmund | semantic average accuracy (%): 86.6 -> 87.6 | Improved |
| add semantic relabeling with shape priors to geometric refinement | mesh refinement from semantic 3D reconstruction input · Dortmund | semantic overall accuracy (%): 92.4 -> 92.7 | No clear effect |
| add semantic relabeling with shape priors to geometric refinement | mesh refinement from semantic 3D reconstruction input · Southbuilding (terrestrial) | semantic average accuracy (%): 78.7 -> 94.5 | Improved |
| add semantic relabeling with shape priors to geometric refinement | mesh refinement from semantic 3D reconstruction input · Southbuilding (terrestrial) | semantic overall accuracy (%): 93.8 -> 98.0 | Improved |
| apply purely geometric baseline refinement to the semantic 3D reconstruction input | mesh refinement from semantic 3D reconstruction input · SynthCity3 A | mean distance to ground truth (m): 0.52 -> 0.44 | Improved |
| apply purely geometric baseline refinement to the semantic 3D reconstruction input | mesh refinement from semantic 3D reconstruction input · Southbuilding (terrestrial) | semantic average accuracy (%): 81.9 -> 78.7 | Hurt |
Closest papers by meaning
- Large-Scale Semantic 3D Reconstruction: An Adaptive Multi-Resolution Model for Multi-Class Volumetric Labeling (CVPR 2016 · similarity 0.91)
- Geometry and Uncertainty-Aware 3D Point Cloud Class-Incremental Semantic Segmentation (CVPR 2023 · similarity 0.89)
- Geometry-Guided Semantic Reconstruction for 3D Instance Segmentation (NeurIPS 2026 · similarity 0.88)
- Diffusion 3D Features (Diff3F): Decorating Untextured Shapes with Distilled Semantic Features (CVPR 2024 · similarity 0.87)
- BFANet: Revisiting 3D Semantic Segmentation with Boundary Feature Analysis (CVPR 2025 · similarity 0.87)
- MESC-3D:Mining Effective Semantic Cues for 3D Reconstruction from a Single Image (CVPR 2025 · 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.
