Research map / 3D vision / Depth and stereo
Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volume
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 |
|---|---|---|---|
| Replace variance-based similarity with average group-wise correlation (ablation) | Base · Tanks and Temples training datasets | Mean F1: 22.16 -> 22.67 | No clear effect |
| Replace variance-based similarity with average group-wise correlation (ablation) | Base · Tanks and Temples training datasets | GPU memory: 11.1 GB -> 6.5 GB | Helped |
| Replace variance-based similarity with average group-wise correlation (ablation) | Base · Tanks and Temples training datasets | time: 2.44 s -> 1.90 s | Helped |
| Use inverse depth regression instead of depth regression (ablation) | AGC · Tanks and Temples training datasets | Mean F1: 22.67 -> 46.96 | Helped |
| Add cascade 3D U-Net filtering (ablation) | AGC-IDR · Tanks and Temples training datasets | Mean F1: 46.96 -> 48.66 | Helped |
| Increase depth sampling number from 192 to 256 (ablation) | CIDER · Tanks and Temples training datasets | Mean F1: 48.66% -> 49.60% | Helped |
| Use CIDER instead of R-MVSNet | MVS methods · DTU evaluation test | Overall: 0.422 -> 0.427 | Mixed |
| Use CIDER instead of R-MVSNet | MVS methods · Tanks and Temples Intermediate | F1: 48.40 -> 46.76 | Hurt |
| Use CIDER instead of R-MVSNet | MVS methods · Tanks and Temples Advanced | F1: 24.91 -> 23.12 | Hurt |
Closest papers by meaning
- Cost Volume Pyramid Based Depth Inference for Multi-View Stereo (CVPR 2020 · similarity 0.93)
- Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching (CVPR 2020 · similarity 0.91)
- Rethinking Depth Estimation for Multi-View Stereo: A Unified Representation (CVPR 2022 · similarity 0.90)
- Non-Parametric Depth Distribution Modelling Based Depth Inference for Multi-View Stereo (CVPR 2022 · similarity 0.90)
- RIAV-MVS: Recurrent-Indexing an Asymmetric Volume for Multi-View Stereo (CVPR 2023 · similarity 0.90)
- Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation (AAAI 2023 · similarity 0.90)
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.