Research map / 3D vision / Depth and stereo
Progressive-X: Efficient, Anytime, Multi-Model Fitting Algorithm
Daniel Barath, Jiri Matas
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 blind generation of many initial hypotheses with progressive proposal-and-consolidation (Prog-X) versus Multi-X | multi-model fitting · AdelaideRMF two-view motion (18 scenes) | misclassification error avg (%): 17.13 -> 10.73 | Improved |
| Prog-X versus Multi-X on homography fitting | multi-model fitting · AdelaideRMF homography (19 scenes) | misclassification error avg (%): 8.71 -> 6.86 | Improved |
| Prog-X versus Multi-X on motion segmentation | multi-model fitting · Hopkins motion segmentation (155 scenes) | misclassification error avg (%): 12.96 -> 8.41 | Improved |
| progressive proposals instead of an initial pool on plane and cylinder fitting with many points | Multi-X · Multi-X plane and cylinder (8 scenes) | processing time (s): 1 407.39 -> 9.36 | Improved |
| progressive fitting versus Multi-X on two-view motion speed | Multi-X · AdelaideRMF two-view motion (18 scenes) | processing time (s): 1.52 -> 14.38 | Hurt |
| Prog-X versus Multi-X on synthetic line fitting | Multi-X · synthetic 2D line fitting, all three scenes (worst of 5 runs) | false positive instances: 6 -> 0 | Improved |
| progressive instance addition versus 2200 initial instances in PEARL: total time on star11 | PEARL · star11 any-time test, total processing time | processing time (ms): 9 880 -> 1 534 | Improved |
| progressive instance addition versus Multi-X on star11 total time | Multi-X · star11 any-time test, total processing time | processing time (ms): 2 065 -> 1 534 | Improved |
| progressive method versus RPA on Hopkins motion segmentation | RPA · Hopkins motion segmentation (155 scenes) | misclassification error avg (%): 9.16 -> 8.41 | Improved |
Closest papers by meaning
- PARSAC: Accelerating Robust Multi-Model Fitting with Parallel Sample Consensus (AAAI 2024 · similarity 0.85)
- Improving Progressive Generation with Decomposable Flow Matching (NeurIPS 2025 · similarity 0.82)
- DGS-LRM: Real-Time Deformable 3D Gaussian Reconstruction From Monocular Videos (NeurIPS 2025 · similarity 0.82)
- Provably Robust Score-Based Diffusion Posterior Sampling for Plug-and-Play Image Reconstruction (NeurIPS 2024 · similarity 0.82)
- Beyond Scores: Proximal Diffusion Models (NeurIPS 2025 · similarity 0.82)
- Floxels: Fast Unsupervised Voxel Based Scene Flow Estimation (CVPR 2025 · similarity 0.82)
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.
