Research map / 3D vision / Depth and stereo
A Confidence-Based Iterative Solver of Depths and Surface Normals for Deep Multi-View Stereo
Wang Zhao, Shaohui Liu, Yi Wei, Hengkai Guo, Yong-Jin Liu
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 solver as post-processing (10 iters) (ablation) | baseline network without joint training · ScanNet test (ablation) | Abs Rel: 0.0735 -> 0.0720 | Improved |
| add end-to-end joint training with solver (no post) (ablation) | depth-normal network · ScanNet test (ablation) | Abs Rel: 0.0735 -> 0.0711 | Improved |
| add solver post-processing after joint training (ablation) | jointly trained network · ScanNet test (ablation) | Abs Rel: 0.0711 -> 0.0665 | Improved |
| add geometric confidence over deep confidence only (ablation) | solver at inference · ScanNet test (ablation) | Abs Rel: 0.0692 -> 0.0667 | Improved |
| add deep confidence to geometric confidence (ablation) | solver at inference · ScanNet test (ablation) | Abs Rel: 0.0667 -> 0.0665 | No clear effect |
| use groundtruth confidence instead of hybrid (ablation) | solver at inference · ScanNet test (ablation) | Abs Rel: 0.0665 -> 0.0599 | Improved |
| increase inference iterations from 1 to 25 (ablation) | solver, 1 view · ScanNet test (ablation) | Abs Rel: 0.0694 -> 0.0670 | Improved |
| increase reference views from 1 to 4 (ablation) | solver, 25 iters · ScanNet test (ablation) | Abs Rel: 0.0670 -> 0.0655 | Improved |
| increase inference iterations from 1 to 10 (ablation) | solver, 2 views · ScanNet test (ablation) | Abs Rel: 0.0690 -> 0.0665 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- DELTAS — DELTAS: Depth Estimation by Learning Triangulation And densification of Sparse points (ECCV 2020) · better on 4 of 4 shared results
- DPSNet — DPSNet: End-to-end Deep Plane Sweep Stereo (ICLR 2019) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Normal Assisted Stereo Depth Estimation (CVPR 2020 · similarity 0.91)
- Fast-MVSNet: Sparse-to-Dense Multi-View Stereo With Learned Propagation and Gauss-Newton Refinement (CVPR 2020 · similarity 0.91)
- Constraining Depth Map Geometry for Multi-View Stereo: A Dual-Depth Approach with Saddle-shaped Depth Cells (ICCV 2023 · similarity 0.91)
- Just a Few Points Are All You Need for Multi-View Stereo: A Novel Semi-Supervised Learning Method for Multi-View Stereo (ICCV 2021 · similarity 0.91)
- Generalized Binary Search Network for Highly-Efficient Multi-View Stereo (CVPR 2022 · similarity 0.91)
- P-MVSNet: Learning Patch-Wise Matching Confidence Aggregation for Multi-View Stereo (ICCV 2019 · similarity 0.91)
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.
