Research map / 3D vision / Depth and stereo
Active Stereo in the Wild through Virtual Pattern Projection
Luca Bartolomei, Matteo Poggi, Fabio Tosi, Andrea Conti, Stefano Mattoccia
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 virtual pattern projection (VPP) without retraining | RAFT-Stereo · Midd-14 | >2 error rate: 15.66 -> 5.09 | Improved |
| add VPP without retraining | rSGM · Midd-14 | >2 error rate: 32.00 -> 10.31 | Improved |
| add VPP without retraining | PSMNet · Midd-14 | >2 error rate: 31.11 -> 15.75 | Improved |
| add Guided Stereo (gd) instead of VPP | PSMNet · Midd-14 | >2 error rate: 31.11 -> 30.66 | No clear effect |
| add VPP on top of Guided Stereo (ablation) | RAFT-Stereo · Midd-14 | >2 error rate: 5.09 -> 6.12 | Hurt |
| add Guided Stereo on top of VPP (ablation) | rSGM · Midd-14 | >2 error rate: 10.31 -> 9.98 | Improved |
| add VPP | RAFT-Stereo · KITTI 142 | >2 error rate: 8.74 -> 6.35 | Improved |
| add 3x3 fixed random patch (ablation) | rSGM · Midd-A | >2 error rate: 26.4 -> 11.9 | Improved |
| use 7x7 bilateral-based adaptive patch instead of fixed patch (ablation) | rSGM · Midd-A | >2 error rate: 11.9 -> 11.0 | Improved |
| use distance-based patches instead of fixed (ablation) | rSGM · Midd-A | >2 error rate: 11.9 -> 12.6 | Hurt |
| combine distance-based and adaptive patches (ablation) | rSGM · Midd-A | >2 error rate: 11.0 -> 12.4 | Hurt |
| add VPP fine-tuning (-ft) versus no fine-tuning (ablation) | RAFT-Stereo · Midd-14 | >2 error rate: 5.09 -> 4.16 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PSMNet — PSMNet: Position-Aware Stereo Merging Network for Room Layout Estimation (CVPR 2022) · better on 1 of 1 shared result
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
- Active Stereo Without Pattern Projector (ICCV 2023 · similarity 0.94)
- Toward Ultrafast Depth Sensing via Active Event-Based Stereo Vision (TPAMI 2026 · similarity 0.87)
- Active Event-based Stereo Vision (CVPR 2025 · similarity 0.87)
- UltraStereo: Efficient Learning-Based Matching for Active Stereo Systems (CVPR 2017 · similarity 0.87)
- ActiveStereoNet: End-to-End Self-Supervised Learning for Active Stereo Systems (ECCV 2018 · similarity 0.86)
- LoS: Local Structure-Guided Stereo Matching (CVPR 2024 · similarity 0.84)
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.
