Research map / 3D vision / Face and shape
Model-free Consensus Maximization for Non-Rigid Shapes
Thomas Probst, Ajad Chhatkuli, Danda Pani Paudel, Luc Van Gool
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 |
|---|---|---|---|
| use exact integer-program solution via BnB instead of the relaxed real-valued version (ablation) | consensus maximizer · Template-to-image, Kinect Paper | inliers detected: 99 / 142 -> 114 / 142 | Improved |
| use exact integer-program solution instead of relaxed (ablation) | consensus maximizer · Template-to-image, T-shirt | inliers detected: 291 / 351 -> 309 / 351 | Improved |
| use exact integer-program solution instead of relaxed (ablation) | consensus maximizer · Template-to-image, Sintel | inliers detected: 44 / 68 -> 53 / 68 | Improved |
| use exact BnB instead of relaxed under high synthetic outlier ratios (ablation) | consensus maximizer · Mocap cloth data, outlier removal | outlier detection: relaxed breaks down beyond 50% outliers; exact works at 80% outliers | Improved |
| use exact solution instead of relaxed (cost in runtime) (ablation) | consensus maximizer · Template-to-image, Kinect Paper | time (s): 5.56 -> 7.59 | Hurt |
| use the graph-based consensus maximizer instead of local-filtering voting | consensus maximizer · Template-to-image, Kinect Paper | inliers detected: 46 / 142 -> 114 / 142 | Improved |
| use the exact method instead of featds | consensus maximizer · Template-to-image, Kinect Paper | inliers detected: 76 / 142 -> 114 / 142 | Improved |
| use the exact method instead of laplacian | consensus maximizer · Template-to-image, Kinect Paper | inliers detected: 114 / 142 -> 126 / 142 | Hurt |
| use the exact method instead of laplacian | consensus maximizer · Template-to-image, T-shirt | inliers detected: 301 / 351 -> 309 / 351 | Improved |
| use the exact method instead of laplacian | consensus maximizer · Template-to-image, Sintel | inliers detected: 44 / 68 -> 53 / 68 | Improved |
| use the exact method instead of maxpoolm | consensus maximizer · Template-to-image, T-shirt | time (s): 608.55 -> 9.66 | Improved |
| cluster matches with k-means and solve each cluster separately (ablation) | exact integer program · dense point surfaces | execution time | Improved |
Closest papers by meaning
- Unsupervised Learning of Consensus Maximization for 3D Vision Problems (CVPR 2019 · similarity 0.92)
- Accelerating Globally Optimal Consensus Maximization in Geometric Vision (TPAMI 2024 · similarity 0.90)
- Robust fitting in computer vision: easy or hard? (ECCV 2018 · similarity 0.89)
- Deterministic Approximate Methods for Maximum Consensus Robust Fitting (TPAMI 2019 · similarity 0.89)
- Guaranteed Outlier Removal With Mixed Integer Linear Programs (CVPR 2016 · similarity 0.89)
- Efficient Globally Optimal Consensus Maximisation with Tree Search (TPAMI 2016 · similarity 0.88)
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.
