Research map / Learning theory and algorithms / Clustering
The Mutex Watershed and its Objective: Efficient, Parameter-Free Graph Partitioning
Steffen Wolf, Alberto Bailoni, Constantin Pape, Nasim Rahaman, Anna Kreshuk, Ullrich Köthe, Fred A. Hamprecht
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 Mutex Watershed instead of multicut (MC-FULL) on same CNN output | CNN affinities · ISBI 2012 (same CNN predictions) | Rand-Score: 0.98029 -> 0.98792 | Improved |
| use Mutex Watershed instead of multicut (MC-FULL) | CNN affinities · ISBI 2012 (same CNN predictions) | runtime (s): 9415.8 -> 43.3 | Improved |
| use Mutex Watershed instead of lifted multicut (LMC) | CNN affinities · ISBI 2012 (same CNN predictions) | Rand-Score: 0.97990 -> 0.98792 | Improved |
| use Mutex Watershed instead of lifted multicut (LMC) | CNN affinities · ISBI 2012 (same CNN predictions) | runtime (s): 966.0 -> 43.3 | Improved |
| add long-range edges to the multicut problem (MC-LOCAL to MC-FULL) (ablation) | Multicut · ISBI 2012 (same CNN predictions) | Rand-Score: 0.70990 -> 0.98029 | Improved |
| use Mutex Watershed with repulsive long-range cues instead of thresholding boundary maps | CNN affinities · ISBI 2012 (same CNN predictions) | Rand-Score: 0.91435 -> 0.98792 | Improved |
| use Mutex Watershed instead of distance-transform watershed (WSDT) | CNN affinities · ISBI 2012 (same CNN predictions) | Rand-Score: 0.88336 -> 0.98792 | Improved |
| use Mutex Watershed instead of watershed seeded at local minima (WS) | CNN affinities · ISBI 2012 (same CNN predictions) | Rand-Score: 0.63958 -> 0.98792 | Improved |
| use Mutex Watershed on CNN output instead of simple thresholding | CNN affinities · ISBI 2012 (same CNN predictions) | runtime (s): 0.2 -> 43.3 | Hurt |
| use Mutex Watershed instead of ResNet + LMC pipeline for top leaderboard entry | ISBI 2012 leaderboard · ISBI 2012 leaderboard | Rand-Score: 0.98788 -> 0.98792 | Improved |
| use Mutex Watershed instead of SCN + LMC | ISBI 2012 leaderboard · ISBI 2012 leaderboard | VI-Score: 0.99144 -> 0.99183 | Improved |
Closest papers by meaning
- The Mutex Watershed: Efficient, Parameter-Free Image Partitioning (ECCV 2018 · similarity 0.98)
- Compassionately Conservative Balanced Cuts for Image Segmentation (CVPR 2018 · similarity 0.88)
- Uniform Hypergraph Partitioning: Provable Tensor Methods and Sampling Techniques (JMLR 2017 · similarity 0.87)
- GASP, a Generalized Framework for Agglomerative Clustering of Signed Graphs and Its Application to Instance Segmentation (CVPR 2022 · similarity 0.87)
- Probabilistic Watershed: Sampling all spanning forests for seeded segmentation and semi-supervised learning (NeurIPS 2019 · similarity 0.86)
- Structural Pruning via Spatial-aware Information Redundancy for Semantic Segmentation (AAAI 2025 · similarity 0.83)
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.
