Research map / Learning theory and algorithms / Clustering
Compassionately Conservative Balanced Cuts for Image Segmentation
Read the paperCheck 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 CCN1-H instead of NCut-H | Hierarchical segmentation · BSDS500 test, Lab affinity, fixed | Covering: 0.51 -> 0.57 | Helped |
| Use CCN1-H instead of NCut-H | Hierarchical segmentation · BSDS500 test, Lab affinity, fixed | PRI: 0.78 -> 0.69 | Hurt |
| Use CCN1-H instead of NCut-H | Hierarchical segmentation · BSDS500 test, Lab affinity, fixed | VI: 1.98 -> 1.81 | Helped |
| Use CCN1-H instead of NCut-H | Hierarchical segmentation · BSDS500 test, mPb affinity, fixed | Covering: 0.30 -> 0.38 | Helped |
| Use CCN1-H instead of NCut-H | Hierarchical segmentation · BSDS500 test, mPb affinity, fixed | PRI: 0.83 -> 0.73 | Hurt |
| Use CCN1-H instead of NCut-H | Hierarchical segmentation · BSDS500 test, mPb affinity, fixed | VI: 2.61 -> 2.35 | Helped |
| Use CCN1-H instead of NCut-H | Hierarchical segmentation · BSDS500 test, PMI affinity, fixed | Covering: 0.37 -> 0.47 | Helped |
| Use CCN1-H instead of NCut-H | Hierarchical segmentation · BSDS500 test, PMI affinity, fixed | PRI: 0.87 -> 0.78 | Hurt |
| Use CCN1-H instead of NCut-H | Hierarchical segmentation · BSDS500 test, PMI affinity, fixed | VI: 2.37 -> 1.99 | Helped |
| Set the CCB parameter to smaller values than τ=1 (ablation) | CCB segmentation · BSDS500 test | segmentation performance | Hurt |
| Use hierarchical 2-way segmentation instead of simultaneous multiway segmentation | CCB segmentation · BSDS500 test | segmentation performance | Helped |
Closest papers by meaning
- Falcon: Fast Proximal Linearization of Normalized Cuts for Unsupervised Image Segmentation (ICLR 2026 · similarity 0.88)
- A Near-Linear Time Approximation Algorithm for Beyond-Worst-Case Graph Clustering (ICML 2024 · similarity 0.86)
- Continuous Partitioning for Graph-Based Semi-Supervised Learning (NeurIPS 2024 · similarity 0.84)
- Topograph: An Efficient Graph-Based Framework for Strictly Topology Preserving Image Segmentation (ICLR 2025 · similarity 0.84)
- Structural Pruning via Spatial-aware Information Redundancy for Semantic Segmentation (AAAI 2025 · similarity 0.84)
- Multidimensional Fractional Programming for Normalized Cuts (NeurIPS 2024 · 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.