Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
A Bayesian Approach Toward Robust Multidimensional Ellipsoid-Specific Fitting
Mingyang Zhao, Xiaohong Jia, Лей Ма, Yuke Shi, Jingen Jiang, Qizhai Li, Dong-ming Yan A, Tiejun Huang
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 uniform outlier component (w>0) to the likelihood (ablation) | BayFit · 2D ellipse fitting, outlier ratio 1% to 15% | fitting error | Improved |
| use RDOS-based automatic initialization of M and w (k=31) instead of manual w=0.3 (ablation) | BayFit · 2D ellipse, 40% outliers | fitting error: k=31 gives relatively improved fitting vs manual w=0.3 | Improved |
| add epsilon-accelerated EM (ablation) | BayFit · 3D ellipsoid, N=200 to 1000 points | EM iterations to converge | Improved |
| parametric vs random sampling of ellipsoid points Y (ablation) | BayFit · 3D ellipsoid, noise and outliers | fitting accuracy | No clear effect |
| BayFit vs RIX under 60% outliers | 3D ellipsoid fitting · 3D ellipsoid, outliers 60% | offset bias Ec: 13.39 -> 0.52 | Improved |
| BayFit vs Huber under 60% outliers | 3D ellipsoid fitting · 3D ellipsoid, outliers 60% | shape deviation Ea: 0.28 -> 0.05 | Improved |
| BayFit vs RIX under 5% noise | 3D ellipsoid fitting · 3D ellipsoid, Gaussian noise 5% | offset bias Ec: 0.67 -> 1.03 | Hurt |
| BayFit vs RIX under 25% noise | 3D ellipsoid fitting · 3D ellipsoid, Gaussian noise 25% | offset bias Ec: 2.68 -> 2.16 | Improved |
| BayFit under occlusion above 50% compared with other fitters | 2D ellipse fitting · 2D ellipse, occlusion 10% to 60% | fitting error | No clear effect |
Closest papers by meaning
- Fast Multidimensional Ellipsoid-Specific Fitting by Alternating Direction Method of Multipliers (TPAMI 2016 · similarity 0.90)
- Filtered Conformal Ellipsoids for Graph-Native Time Series (NeurIPS 2026 · similarity 0.83)
- John Ellipsoids via Lazy Updates (NeurIPS 2024 · similarity 0.83)
- Uncertainty Visualization via Low-Dimensional Posterior Projections (CVPR 2024 · similarity 0.82)
- BOIDS: High-Dimensional Bayesian Optimization via Incumbent-Guided Direction Lines and Subspace Embeddings (AAAI 2025 · similarity 0.81)
- Bayes' Rays: Uncertainty Quantification for Neural Radiance Fields (CVPR 2024 · similarity 0.81)
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.
