AI Paper Atlas

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

Hard-Aware Point-to-Set Deep Metric for Person Re-identification

Rui Yu, Zhiyong Dou, Song Bai, Zhaoxiang Zhang, Yongchao Xu, Xiang Bai

ECCV 2018 · Label-efficient and robust learning › Recognition and long-tail · approach: Supervised person re-identification · 147 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
replace hardest-sample selection with exponential soft hard-aware weighting (sigma=0.5)Hard Triplet · Market-1501mAP: 67.22 -> 69.76Improved
replace hardest-sample selection with polynomial soft hard-aware weighting (alpha=10)Hard Triplet · Market-1501mAP: 67.22 -> 69.43Improved
use uniform weights over the set (sigma -> infinity) instead of hard-aware weights (ablation)HAP2S E · Market-1501mAP: 69.76 -> 1.36Hurt
use uniform weights (alpha=0) instead of hard-aware polynomial weights (ablation)HAP2S P · Market-1501mAP: 69.43 -> 3.11Hurt
increase sigma from 0.5 to 8 (ablation)HAP2S E · Market-1501mAP: 69.76 -> 62.30Hurt
lower alpha from 10 to 4 (ablation)HAP2S P · Market-1501mAP: 69.43 -> 59.07Hurt
swap exponential for polynomial weighting (ablation)HAP2S · Market-1501rank-1: 84.20 -> 84.59No clear effect
add 1000 false-positive outliers to the training set (ablation)HAP2S E · Market-1501mAP: 64.48Hurt
add more than 200 outliers to the training set (ablation)Hard Triplet · Market-1501training stabilityHurt
add re-ranking at test timeHAP2S · Market-1501 (single query)mAP / rank-1: 79.91% mAP and 85.72% rank-1Improved
use HAP2S loss instead of Hard Triplet lossResNet-50 · CUHK03 (labeled)rank-1: 87.65 -> 90.22Improved
use HAP2S E instead of Angular Loss on generic metric learningResNet-50 · CUB-200-2011Recall@1: 54.70 -> 56.08Improved

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