Research map / 3D vision / Depth and stereo
HyperDepth: Learning Depth From Structured Light Without Matching
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 |
|---|---|---|---|
| Use per-scanline random-forest classification followed by regression instead of stereo matching | Structured-light depth estimation · 1.3MP input/output images | speed: 1KHz | Helped |
| Increase tree depth beyond 14 levels (ablation) | HyperDepth · 1000 independently acquired test images | disparity error: 0.005 pixels | Helped |
| Use more than 4 trees per scanline (ablation) | HyperDepth · 1000 independently acquired test images | disparity error: 0.001 in disparity error | Helped |
| Use 4 trees and depth 15 rather than the 3-tree, depth-12 configuration (ablation) | HyperDepth · 1.3 megapixel images | runtime: 2.5msec | Hurt |
| Use the ensemble of multiple trees per line | HyperDepth · Planar targets at distances between 20cm and 350cm | depth error | Helped |
| Use HyperDepth instead of KinectV1 | Structured-light depth estimation · 3D scanning of multiple objects | RMSE | Helped |
| Use regression to recover output disparity rather than explicitly testing subpixel disparities | HyperDepth · Objects at 2.5m | disparity quantization | Helped |
Closest papers by meaning
- UltraStereo: Efficient Learning-Based Matching for Active Stereo Systems (CVPR 2017 · similarity 0.91)
- A Simple yet Universal Framework for Depth Completion (NeurIPS 2024 · similarity 0.89)
- Depth2Pose: A Pose-Based Benchmark for Monocular Depth Estimation without Ground-Truth Depth (NeurIPS 2026 · similarity 0.87)
- TurboSL: Dense Accurate and Fast 3D by Neural Inverse Structured Light (CVPR 2024 · similarity 0.87)
- LoS: Local Structure-Guided Stereo Matching (CVPR 2024 · similarity 0.87)
- MonSter: Marry Monodepth to Stereo Unleashes Power (CVPR 2025 · similarity 0.87)
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.