Research map / Learning theory and algorithms / Matrix and tensor methods
Automatic Solver Generator for Systems of Laurent Polynomial Equations
Evgeniy Martyushev, Snehal Bhayani, Tomáš Pajdla
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 Laurent-polynomial template generator instead of greedy generator | Template solver · Unsynch. rel. pose (#11) | runtime (ms): 0.6 -> 0.3 | Improved |
| use Laurent-polynomial template generator instead of greedy generator | Template solver · Rel. pose lambda1+E+lambda2 7pt (#21) | runtime (ms): 27.9 -> 4.2 | Improved |
| use Laurent-polynomial template generator instead of greedy generator | Template solver · Optimal 3-view triangulation (#35) | runtime (ms): 5.7 -> 1.1 | Improved |
| use Laurent-polynomial template generator instead of greedy generator | Template solver · Time-of-Arrival (4,6) (#31) | runtime (ms): 17.1 -> 5.2 | Improved |
| use Laurent-polynomial template generator instead of greedy generator | Template solver · Optimal PnP Cayley (#12) | runtime (ms): 0.9 -> 1.1 | Hurt |
| use Laurent-polynomial template generator instead of greedy generator | Template solver · Gen. rel. pose + scale 7pt (#24) | runtime (ms): 9.2 -> 9.7 | Hurt |
| use Laurent-polynomial template generator instead of greedy generator | Template solver · Weak PnP (#26) | mean numerical error (log10): -12.36 -> -10.32 | Hurt |
| use our generated solver instead of STD solver | Optimal 3-view triangulation · synthetic | time per call (ms): 1.54 -> 1.34 | Improved |
| use our generated solver instead of QR solver | Optimal 3-view triangulation · synthetic | failure rate error>1 (%): 0.29% -> 1.11% | Hurt |
| use depth formulation instead of homography formulation (Groebner basis) | H13f solver · synthetic | time per call (ms): 27.41 -> 1.10 | Improved |
| use depth formulation instead of homography formulation (resultant) | H13f solver · synthetic | failure rate error>1 (%): 0.06% -> 0.02% | Mixed |
| use our ToA (5,5) solver instead of prior solver | ToA solver · synthetic | time per call (ms): 33.55 -> 18.68 | Improved |
Closest papers by meaning
- Efficient Solvers for Minimal Problems by Syzygy-Based Reduction (CVPR 2017 · similarity 0.90)
- Learning Polynomial Problems with $SL(2, \mathbb{R})$-Equivariance (ICLR 2024 · similarity 0.85)
- Computational Algebra with Attention: Transformer Oracles for Border Basis Algorithms (NeurIPS 2025 · similarity 0.83)
- Learning to compute Gröbner bases (NeurIPS 2024 · similarity 0.83)
- Simulator HC: Regression-based Online Simulation of Starting Problem-Solution Pairs for Homotopy Continuation in Geometric Vision (CVPR 2025 · similarity 0.83)
- Fast Tree-Field Integrators: From Low Displacement Rank to Topological Transformers (NeurIPS 2024 · 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.
