Research map / 3D vision / Depth and stereo
PatchmatchNet: Learned Multi-View Patchmatch Stereo
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 |
|---|---|---|---|
| add adaptive propagation (ablation) | PatchmatchNet · DTU evaluation set | Overall (mm): 0.361 -> 0.352 | Helped |
| add adaptive spatial cost aggregation (ablation) | PatchmatchNet · DTU evaluation set | Overall (mm): 0.380 -> 0.352 | Helped |
| add both adaptive propagation and adaptive evaluation (ablation) | PatchmatchNet · DTU evaluation set | Overall (mm): 0.396 -> 0.352 | Helped |
| increase iterations from 1,1,1 to 2,2,1 (ablation) | PatchmatchNet · DTU evaluation set | Overall (mm): 0.362 -> 0.352 | Helped |
| increase iterations from 2,2,1 to 3,3,1 (ablation) | PatchmatchNet · DTU evaluation set | Overall (mm): 0.352 -> 0.351 | Helped |
| increase iterations beyond 3,3,1 to 4,4,1 or 5,5,1 (ablation) | PatchmatchNet · DTU evaluation set | Overall (mm): 0.351 -> 0.351 | No clear effect |
| add pixel-wise view weighting (ablation) | PatchmatchNet · Tanks & Temples Intermediate | F score: 52.46 -> 53.15 | Helped |
| add pixel-wise view weighting (ablation) | PatchmatchNet · ETH3D Training | F1 score: 61.32 -> 64.21 | Helped |
| use robust training with random selection from the ten best source views (ablation) | PatchmatchNet · Tanks & Temples Intermediate | F score: 52.12 -> 53.15 | Helped |
| use robust training with random selection from the ten best source views (ablation) | PatchmatchNet · ETH3D Training | F1 score: 62.57 -> 64.21 | Helped |
| increase input views from 2 to 5 (ablation) | PatchmatchNet · DTU evaluation set | Overall (mm): 0.385 -> 0.352 | Helped |
| increase input views from 5 to 6 (ablation) | PatchmatchNet · DTU evaluation set | Overall (mm): 0.352 -> 0.353 | Hurt |
Closest papers by meaning
- Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo (CVPR 2023 · similarity 0.91)
- Learning Patch Reconstructability for Accelerating Multi-View Stereo (CVPR 2018 · similarity 0.90)
- Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization (AAAI 2025 · similarity 0.87)
- MSP-MVS: Multi-Granularity Segmentation Prior Guided Multi-View Stereo (AAAI 2025 · similarity 0.87)
- Learning Deformable Hypothesis Sampling for Accurate PatchMatch Multi-View Stereo (AAAI 2024 · similarity 0.87)
- DVP-MVS: Synergize Depth-Edge and Visibility Prior for Multi-View Stereo (AAAI 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.