Research map / Probabilistic and scientific modeling / Molecules and proteins
Efficient Coarse-To-Fine PatchMatch for Large Displacement Optical Flow
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 coarse-to-fine matching instead of single-scale PatchMatch (ablation) | CPM · MPI-Sintel training final | AEE | Helped |
| Increase pyramid levels beyond 5 (ablation) | CPM · MPI-Sintel training final | AEE | Hurt |
| Use a search radius below 4 on lower pyramid levels (ablation) | CPM · MPI-Sintel training final | AEE | Hurt |
| Increase the lower-level search radius (ablation) | CPM · MPI-Sintel training final | AEE | Hurt |
| Reduce grid spacing (ablation) | CPM · MPI-Sintel training final | AEE | Helped |
| Set grid spacing to 1 instead of 3 (ablation) | CPM · MPI-Sintel training final | time | Hurt |
| Increase matching iterations beyond 6 (ablation) | CPM · MPI-Sintel training final | AEE | No clear effect |
| Use CPM matching rather than DeepMatching | MPI-Sintel training final with EpicFlow interpolation · MPI-Sintel training final | AEE: 3.774 -> 3.617 | Helped |
| Use CPM matching rather than EpicFlow | Optical flow estimation · KITTI2012 test | Out-Noc3: 7.88% -> 5.79% | Helped |
| Use CPM matching rather than EpicFlow | Optical flow estimation · Middlebury | AEE: 0.393 -> 0.368 | Helped |
Closest papers by meaning
- Full Flow: Optical Flow Estimation By Global Optimization Over Regular Grids (CVPR 2016 · similarity 0.87)
- Robust Interpolation of Correspondences for Large Displacement Optical Flow (CVPR 2017 · similarity 0.87)
- Optical Flow Matching: Reframing Optical Flow as Continuous Transport Dynamics (CVPR 2026 · similarity 0.87)
- Learning Optical Flow From a Few Matches (CVPR 2021 · similarity 0.87)
- Dense Optical Tracking: Connecting the Dots (CVPR 2024 · similarity 0.85)
- Sparse Global Matching for Video Frame Interpolation with Large Motion (CVPR 2024 · similarity 0.84)
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.