AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Polyhedral Conic Classifiers for Visual Object Detection and Classification

Hakan Cevikalp, Bill Triggs

CVPR 2017 · Detection and segmentation › Open-set and few-shot detection · approach: Open-set and unknown object detection · 31 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 EPCC instead of linear SVMvisual object detection · FDDBAverage Precision (%): 37.6 -> 71.9Improved
Use EPCC instead of linear SVMface detection · ESOGU FacesAverage Precision (%): 47.7 -> 89.1Improved
Use EPCC instead of linear SVMpedestrian detector · INRIA PersonAverage Precision (%): 80.4 -> 85.6Improved
Use OC-EPCC instead of linear SVMvisual classifier · PASCAL VOC 2007Average Precision (%): 68.6 -> 72.5Improved
Use EPCC instead of linear SVMmulticlass classifier · Multiple Featuresclassification accuracy (%): 93.9 ± 1.1 -> 96.3 ± 1.2Improved
Use EPCC instead of linear SVMmulticlass classifier · Letter Recognitionclassification accuracy (%): 59.8 ± 1.7 -> 76.0 ± 1.2Improved
Use EPCC instead of linear SVMopen-set classifier · USPS DigitsAverage Precision (%): 61.9 -> 80.6Improved
Use OC-EPCC instead of EPCCopen-set classifier · LeopardAverage Precision (%): 69.8 ± 7.9 -> 76.6 ± 2.9Improved
Use OC-EPCC instead of EPCCopen-set classifier · AirplaneAverage Precision (%): 15.5 ± 2.8 -> 13.6 ± 1.5Hurt
Use EPCC instead of SVMclosed-set visual classifier · Caltech-256classification accuracy (%): 37.6 ± 0.7 -> 40.1 ± 0.6Improved
Use EPCC instead of SVM2D synthetic classifier · 2D synthetic test setAverage Precision (%): 22.85 -> 86.62Improved
Use polyhedral acceptance regions rather than a half-spacepositive-class classifier · FDDBAverage Precision (%): 37.6 -> 71.9Improved

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