Research map / 3D vision / Depth and stereo
GeoMVSNet: Learning Multi-View Stereo With Geometry Perception
Read the paperCheck 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 |
|---|---|---|---|
| Integrate geometry priors from coarse depth maps using the geometry fusion network (ablation) | GeoMVSNet · DTU evaluation | Overall: 0.3323 -> 0.3207 | Helped |
| Add probability volume embedding alone (ablation) | GeoMVSNet baseline · DTU evaluation | Overall: 0.3323 -> 0.3379 | Hurt |
| Combine geometry fusion with probability volume embedding (ablation) | GeoMVSNet · DTU evaluation | Overall: 0.3323 -> 0.3163 | Helped |
| Add frequency domain filtering to fusion and embedding (ablation) | GeoMVSNet · DTU evaluation | Overall: 0.3163 -> 0.3185 | Hurt |
| Add curriculum learning after frequency domain filtering (ablation) | GeoMVSNet · DTU evaluation | Overall: 0.3185 -> 0.3142 | Helped |
| Add depth distribution similarity loss (ablation) | GeoMVSNet · DTU evaluation | Overall: 0.3142 -> 0.3089 | Helped |
| Add the full combination of geometry fusion, probability volume embedding, frequency filtering, curriculum learning, and distribution similarity loss (ablation) | GeoMVSNet · DTU evaluation | Overall: 0.3323 -> 0.2951 | Helped |
| Use GeoMVSNet rather than the strongest listed baseline | DTU evaluation · DTU evaluation | Overall (mm): 0.303 -> 0.295 | Helped |
| Use GeoMVSNet rather than the strongest listed baseline | Tanks and Temples · Tanks and Temples Intermediate | Mean: 65.90 -> 65.89 | No clear effect |
| Use GeoMVSNet rather than the strongest listed baseline | Tanks and Temples · Tanks and Temples Advanced | Mean: 42.97 -> 43.34 | Helped |
Closest papers by meaning
- Multi-View Stereo Representation Revist: Region-Aware MVSNet (CVPR 2023 · similarity 0.90)
- Fast-MVSNet: Sparse-to-Dense Multi-View Stereo With Learned Propagation and Gauss-Newton Refinement (CVPR 2020 · similarity 0.90)
- Geometry-Aware Stereo Matching via Monocular Disparity Distribution Prior and Gradient Enhancement (AAAI 2026 · similarity 0.90)
- Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation (AAAI 2023 · similarity 0.90)
- DeepMVS: Learning Multi-View Stereopsis (CVPR 2018 · similarity 0.90)
- Deep Stereo Using Adaptive Thin Volume Representation With Uncertainty Awareness (CVPR 2020 · similarity 0.89)
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.