Research map / Video / Video segmentation and tracking
Higher-Order Minimum Cost Lifted Multicuts for Motion Segmentation
Margret Keuper
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 higher-order multicut instead of higher-order spectral clustering | non-lifted HO MC · FBMS-59 Set A | f-measure: about 10% | Improved |
| use higher-order multicut instead of higher-order spectral clustering | non-lifted HO MC · FBMS-59 Set B | f-measure: about 3.5% | Improved |
| use higher-order edges instead of pairwise-only multicut | non-lifted HO MC vs pairwise MC · FBMS-59 Set B | f-measure | No clear effect |
| use motion-adaptive graph construction instead of adding all third order edges (HOPMC) (ablation) | AOMC · FBMS-59 Set B | f-measure: about 1.5% | Improved |
| add lifted edges (ablation) | HOPMC and AOMC · FBMS-59 Set A and B | f-measure | Improved |
| add lifted edges (ablation) | HOPMC · FBMS-59 Set B | segmentation quality vs MCe | No clear effect |
| add third order motion edges to pairwise multicut (Lifted AOMC vs MCe) | Lifted AOMC · DAVIS val | Jmean: 0.552 -> 0.565 | Improved |
| add third order motion edges to pairwise multicut (Lifted AOMC vs MCe) | Lifted AOMC · DAVIS val | Fdecay: 0.034 -> 0.052 | Hurt |
| add lifted edges to higher-order problem (ablation) | KLj higher-order solver · synthetic DAVIS flow instances | segmentation at region boundaries | Improved |
Closest papers by meaning
- Higher-Order Multicuts for Geometric Model Fitting and Motion Segmentation (TPAMI 2022 · similarity 0.91)
- Learning Segmentation from Point Trajectories (NeurIPS 2024 · similarity 0.86)
- Clustering with Hypergraphs: The Case for Large Hyperedges (TPAMI 2016 · similarity 0.86)
- MOVE: Motion-Guided Few-Shot Video Object Segmentation (ICCV 2025 · similarity 0.86)
- HEIR: Learning Graph-Based Motion Hierarchies (NeurIPS 2025 · similarity 0.85)
- GaussianCut: Interactive segmentation via graph cut for 3D Gaussian Splatting (NeurIPS 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.
