AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Weakly Supervised Discriminative Feature Learning With State Information for Person Identification

Hong-Xing Yu, Wei-Shi Zheng

CVPR 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Supervised person re-identification · 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 hard WDBR and WFDR with state informationweakly supervised discriminative learning · DukeMTMC-reIDrank-1: 67.1 -> 72.1Improved
Use hard WDBR and WFDR with state informationweakly supervised discriminative learning · DukeMTMC-reIDmAP: 48.0 -> 53.8Improved
Use state-informed method rather than the strongest reported supervised baselineweakly supervised discriminative learning · Multi-PIEaverage recognition rate: 93.5 -> 97.1Improved
Use state-informed method rather than supervised fine-tuningweakly supervised discriminative learning · CFP frontal-profileAccuracy: 95.50(0.98) -> 95.49(0.70)No clear effect
Add hard WDBR without WFDR (ablation)Basic model · DukeMTMC-reIDrank-1: 56.8 -> 69.4Improved
Add WFDR without WDBR (ablation)Basic model · DukeMTMC-reIDrank-1: 56.8 -> 67.7Improved
Add both WDBR and WFDR (ablation)Basic model · DukeMTMC-reIDrank-1: 56.8 -> 72.1Improved
Add soft WDBR without WFDR (ablation)Basic model · Multi-PIEaverage recognition rate: 54.5 -> 97.0Improved
Add WFDR without WDBR (ablation)Basic model · Multi-PIEaverage recognition rate: 54.5 -> 95.7Improved
Add WFDR to soft-WDBR model (ablation)Basic + WDBR (soft) · Multi-PIEaverage recognition rate: 97.0 -> 97.1No clear effect
Use soft rather than hard WDBR (ablation)Full model · Multi-PIEaverage recognition rate: 95.7 -> 97.1Improved
Use soft rather than hard WDBR (ablation)Full model · DukeMTMC-reIDrank-1: 70.3 -> 72.1Improved

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