AI Paper Atlas

Research map / Detection and segmentation / CNN features

Mask-Guided Attention Network for Occluded Pedestrian Detection

Yanwei Pang, Jin Xie, Muhammad Haris Khan, Rao Muhammad Anwer, Fahad Shahbaz Khan, Ling Shao

ICCV 2019 · Detection and segmentation › CNN features · approach: Pedestrian and crowd detection · 234 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
add mask-guided attention branch (Lmask) (ablation)SPD baseline VGG-16 · CityPersons val HOlog-average miss rate (%): 57.0 -> 52.7Improved
add mask-guided attention branch (Lmask) (ablation)SPD baseline VGG-16 · CityPersons val Rlog-average miss rate (%): 13.8 -> 11.9Improved
add occlusion-sensitive loss (Locc) alone (ablation)SPD baseline VGG-16 · CityPersons val HOlog-average miss rate (%): 57.0 -> 55.6Improved
add Locc on top of mask-guided attention (ablation)MGAN · CityPersons val HOlog-average miss rate (%): 52.7 -> 51.7Improved
add Locc on top of mask-guided attention (ablation)MGAN · CityPersons val Rlog-average miss rate (%): 11.9 -> 11.5Improved
use coarse box-level masks instead of dense pixel-wise masks (ablation)MGAN · CityPersons val HOlog-average miss rate (%): 51.7 -> 52.7No clear effect
use coarse box-level masks instead of dense pixel-wise masks (ablation)MGAN · CityPersons val Rlog-average miss rate (%): 11.2 -> 11.9No clear effect
replace CA attention with mask-guided attention (ablation)MGAN without Locc · CityPersons val HOlog-average miss rate (%): 54.5 -> 52.7Improved
replace CA-VBB attention with mask-guided attention (ablation)MGAN without Locc · CityPersons val HOlog-average miss rate (%): 54.1 -> 52.7Improved
add MGAN to baseline for small heavily occluded pedestriansSPD baseline · CityPersons val HO 50-75 pxlog-average miss rate (%): 66.3 -> 61.7Improved
add MGAN to baseline for large heavily occluded pedestriansSPD baseline · CityPersons val HO >125 pxlog-average miss rate (%): 43.1 -> 37.6Improved
replace Bi-Box with MGANdetector VGG-16 · CityPersons val HOlog-average miss rate (%): 44.2 -> 39.4Improved

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