Research map / Optimization / Optimizers
Efficient Linear Programming for Dense CRFs
Check 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 PROX-LP instead of SG-LP_ℓ | Dense CRF inference · MSRC | Ave. E (×10^3): 3335.6 -> 1274.4 | Helped |
| use PROX-LP instead of SG-LP_ℓ | Dense CRF inference · Pascal VOC 2010 | Ave. E (×10^3): 617.1 -> 507.7 | Helped |
| use PROX-LP instead of SG-LP_ℓ | Dense CRF inference · MSRC | Ave. T (s): 13.6 -> 23.5 | Hurt |
| use PROX-LP_acc instead of SG-LP_ℓ | Dense CRF inference · Pascal VOC 2010 | Ave. T (s): 84.4 -> 14.7 | Helped |
| add restriction to relevant labels (ablation) | PROX-LP · MSRC | Ave. T (s): 23.5 -> 6.3 | Helped |
| add restriction to relevant labels (ablation) | PROX-LP · MSRC | Ave. E (×10^3): 1274.4 -> 1189.8 | Helped |
| add restriction to uncertain pixels, after label restriction (ablation) | PROX-LP_ℓ · MSRC | Ave. T (s): 6.3 -> 3.7 | Helped |
| add restriction to uncertain pixels, after label restriction (ablation) | PROX-LP_ℓ · MSRC | Ave. E (×10^3): 1189.8 -> 1340.0 | Hurt |
| add restriction to uncertain pixels, after label restriction (ablation) | PROX-LP_ℓ · Pascal VOC 2010 | Ave. T (s): 22.1 -> 14.7 | Helped |
| add restriction to uncertain pixels, after label restriction (ablation) | PROX-LP_ℓ · Pascal VOC 2010 | Ave. E (×10^3): 502.1 -> 507.7 | Hurt |
| use modified filtering for conditional-gradient computation instead of divide-and-conquer | Dense CRF LP optimization · Pascal VOC 2010 test image | speedup: 10 - 65 times faster | Helped |
Closest papers by meaning
- Soft Self-labeling and Potts Relaxations for Weakly-supervised Segmentation (CVPR 2025 · similarity 0.86)
- Diffusion-Driven Two-Stage Active Learning for Low-Budget Semantic Segmentation (NeurIPS 2025 · similarity 0.85)
- Towards the Uncharted: Density-Descending Feature Perturbation for Semi-supervised Semantic Segmentation (CVPR 2024 · similarity 0.85)
- Improving Semi-Supervised Semantic Segmentation with Sliced-Wasserstein Feature Alignment and Uniformity (CVPR 2025 · similarity 0.85)
- On the Relaxation of Conditional Independence Assumption for Image Segmentation (NeurIPS 2026 · similarity 0.85)
- DuPL: Dual Student with Trustworthy Progressive Learning for Robust Weakly Supervised Semantic Segmentation (CVPR 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.