AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

Optimizing Elimination Templates by Greedy Parameter Search

Evgeniy Martyushev, Jana Vráblíková, Tomas Pajdla

CVPR 2022 · Learning theory and algorithms › Combinatorial and search optimization · approach: Globally optimal geometric optimization · 10 citations (OpenAlex)

Check 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 changedWhereMeasuredEffect
Use the greedy parameter search to construct the templateElimination-template generation · Rel. pose E+f 6pttemplate size: 21 × 30 -> 11 × 20Improved
Use the greedy parameter search to construct the templateElimination-template generation · Abs. pose refractive P5Ptemplate size: 68 × 93 -> 57 × 73Improved
Use the greedy parameter search to construct the templateElimination-template generation · Refractive P6P+focaltemplate size: 636 × 654 -> 126 × 162Improved
Use the greedy parameter search to construct the templateElimination-template generation · Rel. pose fλ+E+fλ 7pttemplate size: 581 × 659 -> 209 × 277Improved
Use a non-standard quotient-ring basis instead of a standard basis (ablation)Greedy search template generator · Rel. pose λ+E+λ 6pttemplate size: 73 × 125 -> 39 × 95Improved
Use a non-standard quotient-ring basis instead of a standard basis (ablation)Greedy search template generator · Opt. pose w dir 3pttemplate size: 397 × 445* -> 385 × 433*Improved
Use a non-standard quotient-ring basis instead of a standard basis (ablation)Greedy search template generator · Rel. pose E+f 6pttemplate size: 11 × 20 -> 11 × 20No clear effect
Try both row-wise and column-wise greedy reduction and select the smaller result (ablation)Parameterized elimination-template reduction · Minimal problems in geometric computer visiontemplate sizeImproved
Use column-wise rather than row-wise greedy reduction (ablation)Parameterized elimination-template reduction · Some minimal problemstemplate sizeMixed
Apply Schur-complement reduction using sparse fixed-coefficient constraints (ablation)Elimination templates for problems marked with * · Minimal problems with sparse fixed-coefficient polynomialstemplate sizeImproved
Compare the greedy parameter-adjustment strategy with syzygy-based reductionElimination-template generator · Several minimal problemstemplate sizeImproved
Run the proposed solver instead of the publicly available SOTA solverRelative pose solver with unknown common focal length and radial distortion · Rel. pose fλ+E+fλ 7ptaverage runtime (ms): 2.9 ms -> 8.5 msHurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy