Research map / Learning theory and algorithms / Combinatorial and search optimization
New Convex Relaxations for MRF Inference With Unknown Graphs
Zhenhua Wang, Tong Liu, Qinfeng Shi, M. Pawan Kumar, Jianhua Zhang
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 |
|---|---|---|---|
| estimate the MRF graph structure instead of fixed tree-structured graph (ablation) | tree-structured MRF · CAD | F1-score: 70.7 -> 74.4 | Improved |
| estimate the MRF graph structure instead of fixed tree-structured graph (ablation) | tree-structured MRF · TVHI | F1-score: 47.9 -> 62.2 | Improved |
| estimate the graph instead of using a known fully connected graph with TRW-S (ablation) | fully connected MRF · TVHI | F1-score: 55.6 -> 62.2 | Improved |
| use QP+CCCP instead of LP-W+B&B | LP-W+B&B · CAD | F1-score: 73.6 -> 74.4 | Improved |
| use QP+CCCP instead of LP-W+B&B | LP-W+B&B · TVHI | F1-score: 61.1 -> 62.2 | Improved |
| use QP+CCCP instead of LP-W+B&B | LP-W+B&B · BIT | F1-score: 54.2 -> 54.4 | Improved |
| use new LP-M relaxation instead of LP-W | LP-W · CAD | F1-score: 73.6 -> 73.9 | Improved |
| use convex QP relaxation instead of LP-W | LP-W · CAD | F1-score: 73.6 -> 71.7 | Hurt |
| add CCCP iterations on top of the convex QP | QP · CAD | F1-score: 71.7 -> 74.4 | Improved |
| add CCCP iterations on top of the convex QP (ablation) | QP · TVHI | F1-score: 59.4 -> 62.2 | Improved |
| use QP+CCCP or LP-M solver instead of Lan's inference | Lan [15] · CAD test set | estimated energy: -9.62 -> -10.28 | Improved |
| use LP-M instead of QP+CCCP for speed | QP+CCCP · CAD | running time: 2.0 × 10−1 -> 3.0 × 10−2 | Improved |
Closest papers by meaning
- Convex Relaxations for Manifold-Valued Markov Random Fields with Approximation Guarantees (ECCV 2024 · similarity 0.87)
- Continuous Relaxation of MAP Inference: A Nonconvex Perspective (CVPR 2018 · similarity 0.86)
- Efficient semidefinite-programming-based inference for binary and multi-class MRFs (NeurIPS 2020 · similarity 0.86)
- Convex Two-Layer Modeling with Latent Structure (NeurIPS 2016 · similarity 0.86)
- Relaxation-Based Preprocessing Techniques for Markov Random Field Inference (CVPR 2016 · similarity 0.86)
- Graph Alignment via Birkhoff Relaxation (NeurIPS 2025 · 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.
