Research map / Detection and segmentation / Open-set and few-shot detection
Consistent estimation of the max-flow problem: Towards unsupervised image segmentation
Ashif Sikandar Iquebal, Satish T.S. Bukkapatnam
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 estimated flow capacities (iterative MAP) instead of mean shift segmentation | continuous max-flow · AM defect segmentation (SEM, unsupervised comparison) | Dice score (%): 58.6 -> 70.56 | Improved |
| use estimated flow capacities instead of mean shift | continuous max-flow · AM defect segmentation (SEM, unsupervised comparison) | Hausdorff measure: 245 -> 32 | Improved |
| compare unsupervised max-flow with supervised EM method on BraTS 2015 | unsupervised max-flow · BraTS 2015 FLAIR (training data) | Dice score (%): 68 -> 75.6 | Improved |
| compare unsupervised max-flow with best supervised CNN on BraTS 2015 | unsupervised max-flow · BraTS 2015 FLAIR (training data) | Dice score (%): 75.6 -> 88 | Hurt |
| perturb fixed source flow capacity s(x) from 0.2 to 0.35 in plain max-flow | continuous max-flow with manual capacities · AM defect image | segmentation quality | Hurt |
Closest papers by meaning
- Flow Stochastic Segmentation Networks (ICCV 2025 · similarity 0.86)
- FlowSDF: Flow Matching for Medical Image Segmentation Using Distance Transforms (IJCV 2025 · similarity 0.86)
- MedFlowSeg: Flow Matching for Medical Image Segmentation with Frequency-Aware Attention (NeurIPS 2026 · similarity 0.86)
- The Impact Label Noise and Choice of Threshold has on Cross-Entropy and Soft-Dice in Image Segmentation (CVPR 2025 · similarity 0.83)
- Factorized Diffusion Architectures for Unsupervised Image Generation and Segmentation (NeurIPS 2024 · similarity 0.83)
- Unsupervised Universal Image Segmentation (CVPR 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.
