Research map / 3D vision / Depth and stereo
Estimating the Fundamental Matrix Without Point Correspondences With Application to Transmission Imaging
Tobias Wurfl, Andre Aichert, Nicole Maass, Frank Dennerlein, Andreas Maier
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 correspondence-free epipolar consistency estimation instead of feature matching with OpenCV | fundamental matrix estimation · Simulated X-ray CT beads | Frobenius error of F: 1.05E-01 -> 5.45E-03 | Improved |
| add correspondence-free estimation instead of feature matching | fundamental matrix estimation · Simulated X-ray CT beads | Epipole relative error: 6.48E-01 -> 2.62E-02 | Improved |
| add proposed method to jittered initial geometry | jittered projection matrices · Simulated X-ray CT beads | Frobenius error of F: 2.11E-02 -> 5.45E-03 | Improved |
| add Geman-McClure style weighting (eps=1E-3) of empty regions (ablation) | proposed cost function · Simulated X-ray CT beads | Frobenius error of F: 1.53E-02 -> 5.45E-03 | Improved |
| add weighting (eps=1E-3) (ablation) | proposed cost function · Simulated X-ray CT beads | Epipole relative error: 2.83E-02 -> 2.62E-02 | Improved |
| use squared L2 distance instead of L3 (ablation) | distance measure d · Simulated X-ray CT beads | Frobenius error of F: 1.39E-02 -> 5.45E-03 | Improved |
| use squared L2 distance instead of L3/2 (ablation) | distance measure d · Simulated X-ray CT beads | Frobenius error of F: 1.25E-02 -> 5.45E-03 | Improved |
| use squared L2 distance instead of L3 or L3/2 (ablation) | distance measure d · Simulated X-ray CT beads | Epipole relative error: 3.72E-02 -> 2.62E-02 | Improved |
Closest papers by meaning
- Linear Fundamental Matrix Estimation from 7 or 5 Points (CVPR 2026 · similarity 0.86)
- Fundamental Matrix Estimation Using Relative Depths (ECCV 2024 · similarity 0.85)
- Five-Point Fundamental Matrix Estimation for Uncalibrated Cameras (CVPR 2018 · similarity 0.85)
- Minimal Perspective Autocalibration (CVPR 2024 · similarity 0.83)
- Robust Self-calibration of Focal Lengths from the Fundamental Matrix (CVPR 2024 · similarity 0.82)
- Camera Resection from Known Line Pencils and a Radially Distorted Scanline (CVPR 2025 · similarity 0.82)
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.
