AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Symmetry reCAPTCHA

Chris Funk, Yanxi Liu

CVPR 2016 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 25 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
compare human raters to machine symmetry detector [25]symmetry detection on COCO photos · 961 COCO imagesprecision/recallImproved
screen images where algorithm [25] can pass as human using automatic GTadmissible image set · COCOdataset size: 1,200 -> 961Mixed
require at least two unique raters per GT cluster (minPts=2, DBSCAN)GT extraction · COCO labelsGT symmetries: 8,146Improved
raise minimum labels needed for a GT symmetryhuman PR curves · 961 imagesprecision vs recallMixed
use symmetry reCAPTCHA with 40 discriminative imagesnew human raters · 118 raterssuccess rate: 96.31% rotation, 95.05% reflection, 95.81% jointImproved
test human results against null of machine performanceone-sample t-test · 118 ratersp-value: p-value = 0Improved
weight reflection distances by axis lengthsymmetry distance Da, Db, Dc · COCO labelsdistance measureImproved

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