Research map / 3D vision / Depth and stereo
Exploiting & Refining Depth Distributions With Triangulation Light Curtains
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 |
|---|---|---|---|
| Fuse Light Curtain measurements into monocular depth estimates | Monocular depth estimation · KITTI validation | Depth RMSE: 3.176 -> 2.458 | Helped |
| Fuse Light Curtain measurements into stereo depth estimates | Stereo depth estimation · KITTI validation | Depth RMSE: 1.781 -> 1.693 | Helped |
| Use a DPV Fusion Network rather than Bayesian inference to fuse DPVs (ablation) | KITTI lidar upsampling · KITTI | RMSE: 1.118 -> 0.702 | Helped |
| Use a stereo pair at time t rather than monocular images at t and t−1 (ablation) | Depth estimation network · KITTI | RMSE: 2.904 -> 1.737 | Helped |
| Increase the soft cross-entropy target width from σgt = 0.05 to σgt = 0.2 (ablation) | Monocular depth estimation · KITTI | RMSE: 3.24 -> 3.16 | Helped |
| Increase the soft cross-entropy target width from σgt = 0.2 to σgt = 0.3 (ablation) | Monocular depth estimation · KITTI | RMSE: 3.16 -> 3.06 | Helped |
| Add left/right depth and RGB consistency losses (ablation) | Monocular depth estimation with σgt = 0.3 · KITTI | RMSE: 3.06 -> 2.93 | Helped |
| Add edge-aware smoothness loss (ablation) | Monocular depth estimation with consistency losses · KITTI | RMSE: 2.93 -> 2.90 | Helped |
| Place more light curtains at finer depth intervals | Light Curtain depth sensing · Real-world scene | RMS: 1.927 -> 1.574 -> 1.284 | Helped |
| Initialize adaptive sensing with an RGB-predicted depth prior rather than a Gaussian prior | Light Curtain adaptive depth sensing · Real light curtain scenario c2 | Uncertainty Field RMSE | Helped |
Closest papers by meaning
- Energy-Efficient Adaptive 3D Sensing (CVPR 2023 · similarity 0.88)
- Neural RGB(r)D Sensing: Depth and Uncertainty From a Video Camera (CVPR 2019 · similarity 0.86)
- Cross-spectral Gated-RGB Stereo Depth Estimation (CVPR 2024 · similarity 0.86)
- DarkVGGT: Seeing Through Darkness Using Thermal Geometry without Daylight Tax (NeurIPS 2026 · similarity 0.83)
- Blurred LiDAR for Sharper 3D: Robust Handheld 3D Scanning with Diffuse LiDAR and RGB (CVPR 2025 · similarity 0.83)
- Robust Depth Enhancement via Polarization Prompt Fusion Tuning (CVPR 2024 · similarity 0.83)
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.