AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Re-weighting and 1-Point RANSAC-Based PnP Solution to Handle Outliers

Haoyin Zhou, Tao Zhang, Jagadeesan Jayender

TPAMI 2018 · Learning theory and algorithms › Prediction and decision losses · approach: Distributionally robust optimization · 59 citations (OpenAlex)

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 changedWhereMeasuredEffect
use R1PPnP soft re-weighting plus 1-point RANSAC instead of explicit outlier rejection (REPPnP) at high outlier ratesREPPnP · Synthetic ordinary 3D with outliersrobustness to outlier percentage: fails beyond 50% -> handles large percentagesImproved
use R1PPnP instead of REPPnP on quasi-singular casesREPPnP · Synthetic quasi-singular with outliersrobustness to outlier percentage: fails at 5% -> handles large percentagesImproved
use R1PPnP instead of RANSAC+P3P/P4P with refinementRANSAC+P3P + OPnP/ASPnP/RPnP · Synthetic with outliersrotation and translation errorImproved
use R1PPnP instead of RANSAC+P3P/P4P at high outlier percentageRANSAC+P3P/P4P · Synthetic with outliersruntimeImproved
use R1PPnP instead of pure RANSAC+P3P when outliers below 30%RANSAC+P3P · DTU real imagesruntimeHurt
use R1PPnP instead of OPnP as refinementOPnP · DTU real imagesaccuracy and runtimeMixed
use R1PPnP core algorithm instead of other PnP methods without outliers, rotationOPnP, SDP · Synthetic outlier-freerotation accuracyNo clear effect
use R1PPnP core on quasi-singular outlier-free translationother PnP methods · Synthetic quasi-singular outlier-freetranslation accuracyHurt
terminate on stagnating inlier count in a 20-iteration window instead of pose change between iterations (ablation)R1PPnP · Synthetic 50% outliersstopping speed and robustnessImproved
constrain det(R)=1 from the start instead of allowing mirror form and flipping lambda (ablation)R1PPnP · Core algorithmlocal minimaImproved
select the control point near the center of the point cloud (ablation)R1PPnP · Core algorithmconvergence speedImproved

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