AI Paper Atlas

Research map / Detection and segmentation / CNN features

Re-ID Driven Localization Refinement for Person Search

Chuchu Han, Jiacheng Ye, Yunshan Zhong, Xin Tan, Chi Zhang, Changxin Gao, Nong Sang

ICCV 2019 · Detection and segmentation › CNN features · approach: Pedestrian and crowd detection · 165 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
supervise detector boxes with re-ID loss via ROI transform layer instead of regression lossFaster R-CNN · CUHK-SYSU (gallery 100)mAP: 92.23 -> 93.09Improved
supervise detector boxes with re-ID loss via ROI transform layerRetinaNet · CUHK-SYSU (gallery 100)mAP: 91.84 -> 92.78Improved
supervise detector boxes with re-ID lossFaster R-CNN · CUHK-SYSU (gallery 100)Rank-1: 93.24 -> 94.24Improved
add classification loss alone to re-ID supervision of the detector (ablation)localization refinement · CUHK-SYSU (gallery 100)mAP: 92.23 -> 92.34No clear effect
add proxy triplet loss alone to re-ID supervision of the detector (ablation)localization refinement · CUHK-SYSU (gallery 100)mAP: 92.23 -> 92.49No clear effect
combine classification loss with proxy triplet loss (ablation)localization refinement · CUHK-SYSU (gallery 100)mAP: 92.49 -> 93.09Improved
add classification loss to proxy triplet loss (ablation)localization refinement · CUHK-SYSU (gallery 100)Rank-1: 93.48 -> 94.24Improved
apply localization refinement with re-ID model pretrained on a different datasetperson search pipeline · CUHK-SYSU cross-dataset (gallery 100)mAP: 61.87 -> 64.15Improved
increase gallery size from 50 to 4000 (with refinement)Ours · CUHK-SYSUmAPMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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