Research map / Detection and segmentation / Open-set and few-shot detection
K-convexity shape priors for segmentation
Hossam Isack, Lena Gorelick, Karin Ng, Olga Veksler, Yuri Boykov
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 |
|---|---|---|---|
| allow convex parts to overlap (k-convexity) instead of disjoint parts | multi-convexity shape prior · 3D liver segmentation, theta=10 deg, lambda=0.1 | F1: 0.39 -> 0.57 | Improved |
| allow convex parts to overlap (k-convexity) instead of predefined regions | multi-convexity shape prior · 3D liver segmentation, theta=10 deg, lambda=0.1 | F1: 0.07 -> 0.57 | Improved |
| k-convexity vs k-disjoint at theta=20 deg | multi-convexity shape prior · 3D liver segmentation, lambda=1 | F1: 0.50 -> 0.73 | Improved |
| k-convexity vs k-regions at theta=45 deg | multi-convexity shape prior · 3D liver segmentation, lambda=0.1 | F1: 0.84 -> 0.83 | No clear effect |
| loosen shape tightness theta from 10 to 45 degrees for k-regions (ablation) | k-regions · 3D liver segmentation, lambda=0.1 | F1: 0.07 -> 0.84 | Improved |
| loosen shape tightness theta from 10 to 45 degrees for k-convexity (ablation) | k-convexity · 3D liver segmentation, lambda=0.1 | F1: 0.57 -> 0.83 | Improved |
| increase smoothness lambda from 0.1 to 1 for k-disjoint at theta=10 (ablation) | k-disjoint · 3D liver segmentation | F1: 0.39 -> 0.29 | Hurt |
| k-convexity vs k-regions on connected components at theta=10, lambda=1 | multi-convexity shape prior · 3D liver segmentation | connected components: 5.83 -> 2.67 | Improved |
| k-convexity vs k-disjoint on connected components at theta=20, lambda=0.1 | multi-convexity shape prior · 3D liver segmentation | connected components: 1.25 -> 1.17 | No clear effect |
| allow overlapping labels with regular (non-submodular) convexity | k-convexity with regular convexity · synthetic proof-of-concept | qualitative | Improved |
Closest papers by meaning
- Convexity Shape Prior for Binary Segmentation (TPAMI 2016 · similarity 0.90)
- D-Convexity: A Unified Differentiable Convex Shape Prior via Quasi-Concavity for Data-driven Image Segmentation (CVPR 2026 · similarity 0.89)
- Convexity Shape Constraints for Image Segmentation (CVPR 2016 · similarity 0.89)
- Convex Shape Prior for Multi-Object Segmentation Using a Single Level Set Function (ICCV 2019 · similarity 0.87)
- Convex Combination Star Shape Prior for Data-driven Image Semantic Segmentation (CVPR 2025 · similarity 0.85)
- Avoiding spurious sharpness minimization broadens applicability of SAM (ICML 2025 · similarity 0.82)
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.
