Research map / Learning theory and algorithms / Prediction and decision losses
Truncated Max-Of-Convex Models
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 TMCM range expansion instead of parsimonious labeling and co-occurrence move-making | Image inpainting and denoising (penguin, house) · Middlebury images | energy | Helped |
| Use TMCM range expansion instead of parsimonious labeling and co-occurrence move-making | Stereo matching · Middlebury stereo: tsukuba, venus, teddy | energy | Helped |
| Use TMCM range expansion instead of the baselines | TMCM · Middlebury stereo: cones | energy | Mixed |
| Use TMCM range expansion instead of parsimonious labeling | Max-of-linear TMCM, clique weight 5 · Synthetic lattices | energy | Mixed |
| Use TMCM range expansion instead of the baselines | Max-of-linear TMCM, clique weight 10 · Synthetic lattices | energy | Helped |
| Use TMCM range expansion instead of co-occurrence move-making | Max-of-linear TMCM, clique weight 5 · Synthetic lattices | energy | Helped |
| Use TMCM range expansion instead of the baselines | TMCM · Synthetic lattices | convergence time | Hurt |
| Use smaller interval lengths (ablation) | TMCM range expansion · Synthetic lattices | energy | No clear effect |
| Set m = 3 rather than m = 1 (ablation) | TMCM range expansion · Image inpainting and denoising (penguin) | visual quality | Helped |
| Set m = 3 rather than m = 1 (ablation) | TMCM range expansion · Stereo matching (tsukuba, venus) | robustness to inaccurate clique definitions | Helped |
Closest papers by meaning
- Min Norm Point Algorithm for Higher Order MRF-MAP Inference (CVPR 2016 · similarity 0.86)
- Learning with Stochastic Orders (ICLR 2023 · similarity 0.85)
- D-Convexity: A Unified Differentiable Convex Shape Prior via Quasi-Concavity for Data-driven Image Segmentation (CVPR 2026 · similarity 0.85)
- Consensus Maximisation Using Influences of Monotone Boolean Functions (CVPR 2021 · similarity 0.85)
- Tilted Sharpness-Aware Minimization (ICML 2025 · similarity 0.85)
- Adaptive and Move Making Auxiliary Cuts for Binary Pairwise Energies (CVPR 2017 · 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.