Research map / 3D vision / Human pose estimation
Relative Pose from SIFT Features
Daniel Barath, Zuzana Kukelova
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 |
|---|---|---|---|
| replace 5PT with SIFT-based solver | GC-RANSAC E estimation · KITTI | run-time (ms): 276.4 -> 53.6 | Improved |
| replace 5PT with SIFT-based solver | GC-RANSAC E estimation · KITTI | rotation error (deg): 2.8 -> 2.8 | No clear effect |
| replace 5PT with SIFT-based solver | GC-RANSAC E estimation · PhotoTourism | run-time (ms): 847.3 -> 108.1 | Improved |
| replace 5PT with SIFT-based solver | GC-RANSAC E estimation · PhotoTourism | translation error (deg): 2.2 -> 2.3 | No clear effect |
| replace 7PT with SIFT-based solver | GC-RANSAC F estimation · KITTI | run-time (ms): 154.4 -> 67.3 | Improved |
| replace 7PT with SIFT-based solver | GC-RANSAC F estimation · PhotoTourism | translation error (deg): 7.8 -> 6.7 | Improved |
| replace 7PT with SIFT-based solver | GC-RANSAC F estimation · PhotoTourism | run-time (ms): 127.3 -> 48.8 | Improved |
| replace 6PT with SIFT-based solver | GC-RANSAC F + focal length · PhotoTourism | rotation error (deg): 2.6 -> 1.5 | Improved |
| replace 6PT with SIFT-based solver | GC-RANSAC F + focal length · PhotoTourism | run-time (ms): 743.0 -> 290.7 | Improved |
| replace 6PT with SIFT-based solver | GC-RANSAC F + focal length · KITTI | run-time (ms): 225.8 -> 61.5 | Improved |
| replace 6PT with SIFT-based solver | GC-RANSAC F + focal length · KITTI | focal length error: 0.80 -> 0.77 | No clear effect |
| use SIFT-based instead of point-based solver in synthetic noise test (E and F) | SIFT-based solvers · synthetic | symmetric epipolar error | Hurt |
Closest papers by meaning
- On Making SIFT Features Affine Covariant (IJCV 2023 · similarity 0.93)
- Optimal Relative Pose With Unknown Correspondences (CVPR 2016 · similarity 0.91)
- Camera Calibration by Global Constraints on the Motion of Silhouettes (ICCV 2017 · similarity 0.88)
- From Correspondences to Pose: Non-minimal Certifiably Optimal Relative Pose without Disambiguation (CVPR 2024 · similarity 0.88)
- Relative Pose Estimation through Affine Corrections of Monocular Depth Priors (CVPR 2025 · similarity 0.85)
- Practical Solutions to the Relative Pose of Three Calibrated Cameras (CVPR 2025 · similarity 0.85)
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.
