Research map / 3D vision / Human pose estimation
BabelCalib: A Universal Approach to Calibrating Central Cameras
Yaroslava Lochman, Kostiantyn Liepieshov, Jianhui Chen, Michal Perdoch, Christopher Zach, James Pritts
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 corner correction via radial fundamental matrix (ablation) | BabelCalib initial estimate · synthetic chessboard, noise sigma 0-2 | median rotation, translation and RMS reprojection error: 31%, 28%, 33% | Improved |
| use division model of degree 4 instead of degree 2 (ablation) | BabelCalib back-projection · Division model selection (refined) | RMS reprojection error [px]: 0.647 -> 0.587 | Improved |
| use division model of degree 6 instead of degree 4 (ablation) | BabelCalib back-projection · Division model selection (refined) | RMS reprojection error [px]: 0.587 -> 0.587 | No clear effect |
| raise division model degree from 4 to 8 (ablation) | BabelCalib back-projection · Division model selection (refined) | RMS reprojection error [px]: 0.587 -> 1.020 | Hurt |
| raise division model degree from 4 to 10 (ablation) | BabelCalib back-projection · Division model selection (refined) | RMS reprojection error [px]: 0.587 -> 4.812 | Hurt |
| replace Kalibr KB calibration with BabelCalib regression | Kalibr-KB · Kalibr dataset KB | failures: 3/8 -> 0/8 | Improved |
| replace OpenCV KB calibration with BabelCalib regression | OpenCV-KB · Kalibr dataset KB | RMS reprojection error [px]: 2.291 -> 0.173 | Improved |
| replace Kalibr UCM calibration with BabelCalib regression | Kalibr-UCM · OV-Cube UCM | RMS reprojection error [px]: 0.440 -> 0.316 | Improved |
| replace Kalibr UCM calibration with BabelCalib regression | Kalibr-UCM · OCamCalib UCM | RMS reprojection error [px]: 0.776 -> 0.784 | No clear effect |
| replace Kalibr DS calibration with BabelCalib regression | Kalibr-DS · Kalibr dataset DS | RMS reprojection error [px]: 0.264 -> 0.326 | Hurt |
| replace OCamCalib DIV with BabelCalib DIV | OCamCalib-DIV · OCamCalib DIV | RMS reprojection error [px] | No clear effect |
Closest papers by meaning
- AnyCalib: On-Manifold Learning for Model-Agnostic Single-View Camera Calibration (ICCV 2025 · similarity 0.87)
- Revisiting Calibration of Wide-Angle Radially Symmetric Cameras (ECCV 2024 · similarity 0.86)
- Why Having 10,000 Parameters in Your Camera Model Is Better Than Twelve (CVPR 2020 · similarity 0.86)
- Inferring Bias and Uncertainty in Camera Calibration (IJCV 2021 · similarity 0.85)
- Scene-agnostic Pose Regression for Visual Localization (CVPR 2025 · similarity 0.84)
- Camera Resection from Known Line Pencils and a Radially Distorted Scanline (CVPR 2025 · 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.
