AI Paper Atlas

Research map / Detection and segmentation / CNN features

Boosted Random Ferns for Object Detection

Michael Villamizar, Juan Andrade‐Cetto, Alberto Sanfeliu, Francesc Moreno-Noguer

TPAMI 2017 · Detection and segmentation › CNN features · approach: Region-based and query-based detectors · 14 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 HOG cell comparisons instead of pixel intensity comparisons as fern features (ablation)BRFs · INRIA HorsesPR-EER: 54.6 -> 74.1Improved
use HOG cell comparisons instead of pixel intensity comparisons (ablation)BRFs · TUD MotorbikesPR-EER: 53.4 -> 84.4Improved
use HOG cell comparisons instead of pixel intensity comparisons (ablation)BRFs · UIUC CarsPR-EER: 94.9 -> 98.9Improved
add boosting to random ferns (25 weak learners, 2 features) (ablation)Random Ferns · UIUC Carsmean EER: 15.9 -> 75.9Improved
add boosting to random ferns (25 weak learners, 2 features) (ablation)Random Ferns · INRIA Horsesmean EER: 14.6 -> 37.4Improved
add boosting to random ferns (25 weak learners, 2 features) (ablation)Random Ferns · TUD Motorbikesmean EER: 18.3 -> 56.3Improved
add boosting to random ferns (300 weak learners, 8 features) (ablation)Random Ferns · UIUC Carsmean EER: 90.2 -> 98.9Improved
share fern parameter sets (25 WLs, 2 features) (ablation)BRFs with independent ferns · UIUC Carsmean EER: 84.9 -> 75.9Hurt
share fern parameter sets (100 WLs, 6 features) (ablation)BRFs with independent ferns · TUD Motorbikesmean EER: 82.1 -> 83.4No clear effect
share fern parameter sets (300 WLs, 8 features) (ablation)BRFs with independent ferns · INRIA Horsesmean EER: 73.9 -> 74.1No clear effect
increase shared fern pool size R from 2 to 10 (ablation)BRFs · UIUC CarsPR-EER and time (ms): 97.8 -> 98.9; 23 -> 76 msMixed
increase shared fern pool size R from 10 to 30 (ablation)BRFs · TUD MotorbikesPR-EER and time (ms): 84.4 -> 84.7; 150 -> 428 msNo clear effect

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