AI Paper Atlas

Research map / Detection and segmentation / CNN features

Socially Constrained Structural Learning for Groups Detection in Crowd

Francesco Solera, Simone Calderara, Rita Cucchiara

TPAMI 2016 · Detection and segmentation › CNN features · approach: Pedestrian and crowd detection · 119 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
learn feature weights with structural SVM instead of random equal weights (ablation)group detection · hotelG-MITRE precision: 71.0 -> 97.3Improved
learn feature weights with structural SVM instead of random equal weights (ablation)group detection · student003G-MITRE precision: 59.9 -> 81.7Improved
learn feature weights with structural SVM instead of random equal weights (ablation)group detection · ethG-MITRE recall: 65.2 -> 94.2Improved
train with G-MITRE loss instead of pairwise loss (ablation)structural SVM group detection · stu003G-MITRE precision: 68.9 -> 80.1Improved
train with G-MITRE loss instead of pairwise loss (ablation)structural SVM group detection · ethG-MITRE precision: 88.7 -> 91.9Improved
train with G-MITRE loss instead of pairwise loss (ablation)structural SVM group detection · hotelG-MITRE precision: 90.1 -> 89.2No clear effect
use learned method instead of Ge et al. clusteringgroup detection · hotelG-MITRE precision: 89.2 -> 97.3Improved
use learned method instead of Ge et al. clusteringgroup detection · student003G-MITRE precision: 77.2 -> 81.7Improved
replace ground-truth trajectories with detector and tracker outputproposed method · hotelG-MITRE precision: 97.3 -> 77.9Hurt
replace ground-truth trajectories with detector and tracker outputproposed method · studentG-MITRE precision: 81.7 -> 75.0Hurt
use proposed method instead of Yamaguchi et al. on automatic trajectoriesgroup detection · hotelG-MITRE precision: 46.3 -> 77.9Improved
use sequential or online training instead of batch (ablation)proposed method · gal1 of GVEII with growing densityF1Improved

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