Research map / 3D vision / Human pose estimation
Correspondences of the Third Kind: Camera Pose Estimation from Object Reflection
Kohei Yamashita, Vincent Lepetit, Ko Nishino
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 reflection correspondences (ablation) | camera pose estimation from two images · nLMVS-Synth test | mean relative rotation error (deg): 6.6 deg -> 4.5 deg | Improved |
| add joint iterative shape and pose estimation (ablation) | camera pose estimation from two images · nLMVS-Synth test | mean relative rotation error (deg): 6.2 deg -> 4.5 deg | Improved |
| add GBR data augmentation when training feature extractors (ablation) | correspondence feature network · nLMVS-Synth test | mean relative rotation error (deg): 7.4 deg -> 4.5 deg | Improved |
| add reflection correspondences on shapes with varied flatness (lambda 0.69 to 1.44) (ablation) | camera pose estimation · GBR-distorted Stanford Bunny | mean pose error | Improved |
| use proposed method instead of LightGlue on real two-view images | camera pose estimation · nLMVS-Real Bunny | camera pose error (deg): 93.6 deg -> 7.2 deg | Improved |
| use proposed method instead of COLMAP on two cropped views | camera pose estimation · nLMVS-Real | camera pose error (deg): 15.7 deg -> 1.9 deg | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- LightGlue — LightGlue: Local Feature Matching at Light Speed (ICCV 2023) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- RefPose: Leveraging Reference Geometric Correspondences for Accurate 6D Pose Estimation of Unseen Objects (CVPR 2025 · similarity 0.87)
- Rethinking Correspondence-based Category-Level Object Pose Estimation (CVPR 2025 · similarity 0.87)
- UNOPose: Unseen Object Pose Estimation with an Unposed RGB-D Reference Image (CVPR 2025 · similarity 0.86)
- From Correspondences to Pose: Non-minimal Certifiably Optimal Relative Pose without Disambiguation (CVPR 2024 · similarity 0.86)
- RUBIK: A Structured Benchmark for Image Matching across Geometric Challenges (CVPR 2025 · similarity 0.86)
- Structure-Aware Correspondence Learning for Relative Pose Estimation (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.
