Research map / 3D vision / Human pose estimation
Object-Based Visual Camera Pose Estimation From Ellipsoidal Model and 3D-Aware Ellipse Prediction
Matthieu Zins, Gilles Simon, Marie‐Odile Berger
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 3D-coherent ellipse prediction instead of inscribed ellipses (ablation) | pose from ellipsoids · 7-Scenes Chess seq 2 | median position error (cm): 11.62 -> 6.46 | Improved |
| add 3D-coherent ellipse prediction instead of inscribed ellipses (ablation) | pose from ellipsoids · 7-Scenes Chess seq 3 | median position error (cm): 10.56 -> 7.03 | Improved |
| add 3D-coherent ellipse prediction instead of inscribed ellipses (ablation) | pose from ellipsoids · 7-Scenes Chess seq 5 | median position error (cm): 10.20 -> 6.42 | Improved |
| add 3D-coherent ellipse prediction instead of inscribed ellipses (ablation) | camera position · LINEMOD driller (ADD 10%) | % correct positions: 42.86 -> 85.71 | Improved |
| add 3D-coherent ellipse prediction instead of inscribed ellipses (ablation) | camera position · LINEMOD driller (reprojection 5 px) | % correct positions: 16.13 -> 96.31 | Improved |
| add 3D-coherent ellipse prediction instead of inscribed ellipses (ablation) | camera position · WatchPose (easy) | median position error (mm): 84.11 -> 26.63 | Improved |
| add 3D-coherent ellipse prediction instead of inscribed ellipses (ablation) | camera position · WatchPose (hard) | median position error (mm): 120.31 -> 54.12 | Improved |
| add predicted ellipses instead of inscribed at small test distance (ablation) | full pose · Synthetic YCB scene test case 1 | mean position error (m): 0.057 -> 0.048 | Improved |
| add predicted ellipses instead of inscribed at large distance (ablation) | full pose · Synthetic YCB scene test case 3 | mean position error (m): 0.096 -> 0.118 | Hurt |
| use sampling-based implicit function loss instead of multi-bin loss (ablation) | ellipse prediction network · 7-Scenes objects (Xbox left) | mean IoU: 0.869 -> 0.943 | Improved |
| use sampling-based loss instead of multi-bin loss (ablation) | ellipse prediction network · 7-Scenes objects (Video Games) | mean IoU: 0.946 -> 0.945 | No clear effect |
| use central matrix diag(alpha, beta) instead of diag(1/alpha^2, 1/beta^2) (ablation) | ellipse embedding function · Synthetic driller ellipse prediction | mean IoU: 0.873 -> 0.921 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- Pix2Pose — Pix2Pose: Pixel-Wise Coordinate Regression of Objects for 6D Pose Estimation (ICCV 2019) · better on 1 of 3 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
- Perspective-1-Ellipsoid: Formulation, Analysis and Solutions of the Camera Pose Estimation Problem from One Ellipse-Ellipsoid Correspondence (IJCV 2023 · similarity 0.92)
- ELLIPSDF: Joint Object Pose and Shape Optimization With a Bi-Level Ellipsoid and Signed Distance Function Description (ICCV 2021 · similarity 0.88)
- LEAP: Liberate Sparse-View 3D Modeling from Camera Poses (ICLR 2024 · similarity 0.85)
- 3D-Aware Hypothesis & Verification for Generalizable Relative Object Pose Estimation (ICLR 2024 · similarity 0.85)
- NOPE: Novel Object Pose Estimation from a Single Image (CVPR 2024 · similarity 0.85)
- UNOPose: Unseen Object Pose Estimation with an Unposed RGB-D Reference Image (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.
