Research map / 3D vision / Depth and stereo
Booster: A Benchmark for Depth From Images of Specular and Transparent Surfaces
Pierluigi Zama Ramirez, Alex Costanzino, Fabio Tosi, Matteo Poggi, Samuele Salti, Stefano Mattoccia, Luigi Di Stefano
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 fine-tuning on the Booster training split (ablation) | CREStereo · Booster Balanced (full-res GT, all pixels) | bad-2: 33.07 -> 29.48 | Improved |
| add fine-tuning on the Booster training split (ablation) | RAFT-Stereo · Booster Balanced (quarter-res GT, all pixels) | bad-2: 17.43 -> 10.72 | Improved |
| add fine-tuning on the Booster training split (ablation) | CFNet · Booster Balanced (full-res GT, all pixels) | bad-2 and MAE: bad-2 61.34 -> 66.94; MAE 27.60 -> 19.65 | Mixed |
| add fine-tuning on the Booster training split (ablation) | RAFT-Stereo · Booster Balanced (full-res GT, material Class 0) | bad-2: 28.14 -> 30.94 | Hurt |
| add fine-tuning on the Booster training split (ablation) | CREStereo · Booster Balanced (full-res GT, material Class 3) | bad-2: 79.50 -> 67.40 | Improved |
| add fine-tuning on the Booster training split (ablation) | CREStereo · Booster Unbalanced (material Class 3) | bad-2: 86.39 -> 63.37 | Improved |
| add fine-tuning on the Booster training split (ablation) | RAFT-Stereo · Booster Unbalanced (full-res GT, all pixels) | bad-2 and MAE: bad-2 54.47 -> 58.55; MAE 9.52 -> 5.68 | Mixed |
| add fine-tuning on the Booster training split (ablation) | CREStereo · Booster Unbalanced (full-res GT, all pixels) | bad-2: 56.00 -> 53.52 | Improved |
| add fine-tuning on the Booster training split (ablation) | DPT · Booster Mono (all pixels) | Abs Rel: 0.069 -> 0.065 | Improved |
| add fine-tuning on the Booster training split (ablation) | MiDaS · Booster Mono (all pixels) | Abs Rel: 0.073 -> 0.073 | No clear effect |
| add Boosting multi-resolution merging | MiDaS · Booster Mono (all pixels) | Abs Rel: 0.073 -> 0.072 | No clear effect |
| add Boosting multi-resolution merging | LeRes · Booster Mono (all pixels) | Abs Rel: 0.077 -> 0.078 | No clear effect |
Closest papers by meaning
- Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data (CVPR 2024 · similarity 0.87)
- Depth2Pose: A Pose-Based Benchmark for Monocular Depth Estimation without Ground-Truth Depth (NeurIPS 2026 · similarity 0.87)
- Depth Pro: Sharp Monocular Metric Depth in Less Than a Second (ICLR 2025 · similarity 0.86)
- PatchFusion: An End-to-End Tile-Based Framework for High-Resolution Monocular Metric Depth Estimation (CVPR 2024 · similarity 0.86)
- Depth Anything V2 (NeurIPS 2024 · similarity 0.86)
- HELVIPAD: A Real-World Dataset for Omnidirectional Stereo Depth Estimation (CVPR 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.
