Research map / Detection and segmentation / Open-set and few-shot detection
Interactive Image Segmentation Using Constrained Dominant Sets
Eyasu Zemene, Marcello Pelillo
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 |
|---|---|---|---|
| select sigma per image instead of self-tuning sigma (ablation) | CDS · GrabCut (scribbles on foreground) | error rate: 3.57 -> 2.72 | Improved |
| select sigma per image instead of one per-dataset sigma (ablation) | CDS · GrabCut (scribbles on foreground) | error rate: 3.80 -> 2.72 | Improved |
| use self-tuning sigma instead of single per-dataset sigma (ablation) | CDS · GrabCut (scribbles on foreground) | error rate: 3.80 -> 3.57 | Improved |
| per-image best sigma instead of self-tuning sigma (ablation) | CDS · Berkeley (100 images) | Jaccard index: 93 -> 95 | Improved |
| self-tuning vs single sigma (ablation) | CDS · Berkeley (100 images) | Jaccard index: 93 -> 93 | No clear effect |
| per-image best sigma instead of single sigma under loose box L=120% (ablation) | CDS · GrabCut 41 images, bounding box looseness L=120% | error rate: 5.9 -> 4.4 | Improved |
| per-image best sigma instead of self-tuning at L=0% (ablation) | CDS · GrabCut 41 images, bounding box looseness L=0% | error rate: 7.54 -> 6.0 | Improved |
| increase box looseness from 0% to 600% | GrabCut · GrabCut 41 images | error rate: 7.4 -> 13.7 | Hurt |
| increase box looseness from 0% to 600% | CDS_Single Sigma · GrabCut 41 images | error rate: 7.48 -> 6.29 | Improved |
| increase box looseness from 0% to 600% | LooseCut · GrabCut 41 images | error rate: 7.9 -> 6.8 | Improved |
| increase fraction of erroneous scribbles from 0% to 100% | CDS · GrabCut synthetic scribbles | error rate | No clear effect |
Closest papers by meaning
- Dominant Sets for “Constrained” Image Segmentation (TPAMI 2018 · similarity 0.96)
- MultiSeg: Semantically Meaningful, Scale-Diverse Segmentations From Minimal User Input (ICCV 2019 · similarity 0.86)
- TETRIS: Towards Exploring the Robustness of Interactive Segmentation (AAAI 2024 · similarity 0.84)
- MFP: Making Full Use of Probability Maps for Interactive Image Segmentation (CVPR 2024 · similarity 0.84)
- Rethinking Interactive Image Segmentation with Low Latency High Quality and Diverse Prompts (CVPR 2024 · similarity 0.84)
- Hierarchy-Agnostic Unsupervised Segmentation: Parsing Semantic Image Structure (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.
