Research map / 3D vision / Depth and stereo
Mesh-Guided Multi-View Stereo With Pyramid Architecture
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 |
|---|---|---|---|
| use mesh guidance to complete and refine coarse-scale depth maps (ablation) | three-scale ACMH MVS · ETH3D validation Avg, 2cm | F1 score: 71.93 -> 78.65 | Helped |
| add MG-Conf to remove unreliable depth estimates (ablation) | MG-MVS · ETH3D validation Avg, 2cm | F1 score: 78.11 -> 78.65 | Helped |
| add an untextured region detector and preserve depths there with geometric consistency (ablation) | MG-MVS · ETH3D validation Avg, 2cm | F1 score: 74.03 -> 78.65 | Helped |
| add ASPP with attention to MG-Conf (ablation) | MG-Conf · MG-Conf validation | error recall rate: 0.5851 -> 0.6919 | Helped |
| add ASPP with attention to MG-Conf (ablation) | MG-MVS · ETH3D validation Avg, 2cm | F1 score: 78.05 -> 78.65 | Helped |
| add ASPP with attention to MG-Conf (ablation) | MG-Conf · MG-Conf validation | prediction accuracy: 0.8609 -> 0.8119 | Hurt |
| use mesh-guided reconstruction | MVS on ETH3D high-resolution test set · ETH3D high-resolution test set, 2cm | completeness: 75.23 -> 87.11 | Helped |
| use mesh-guided reconstruction | MVS on ETH3D high-resolution test set · ETH3D high-resolution test set, 2cm | accuracy: 82.15 -> 80.32 | Hurt |
Closest papers by meaning
- Multi-View Stereo Representation Revist: Region-Aware MVSNet (CVPR 2023 · similarity 0.90)
- Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization (AAAI 2025 · similarity 0.90)
- Fast-MVSNet: Sparse-to-Dense Multi-View Stereo With Learned Propagation and Gauss-Newton Refinement (CVPR 2020 · similarity 0.90)
- Multi-view Reconstruction via SfM-guided Monocular Depth Estimation (CVPR 2025 · similarity 0.88)
- MonSter: Marry Monodepth to Stereo Unleashes Power (CVPR 2025 · similarity 0.88)
- MVSAnywhere: Zero-Shot Multi-View Stereo (CVPR 2025 · similarity 0.87)
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.