AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Generalizing A Person Retrieval Model Hetero- and Homogeneously

Zhun Zhong, Liang Zheng, Shaozi Li, Yi Yang

ECCV 2018 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Geometric alignment and synthetic-to-real adaptation · 508 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 triplet loss on source samples (ablation)Baseline · Duke to Market-1501Rank-1 accuracy: 44.6 -> 48.6Improved
add camera invariance loss using camera-style-transferred target images (ablation)Baseline · Duke to Market-1501Rank-1 accuracy: 44.6 -> 60.6Improved
add domain connectedness (unlabeled target negatives) on top of triplet loss (ablation)Baseline + LT · Duke to Market-1501Rank-1 accuracy: 48.6 -> 49.8Improved
add domain connectedness (unlabeled target negatives) on top of triplet loss (ablation)Baseline + LT · CUHK03 to Market-1501Rank-1 accuracy: 46.1 -> 48.9Improved
add joint camera invariance and domain connectedness over camera invariance alone (ablation)Baseline + LC · CUHK03 to Market-1501Rank-1 accuracy: 53.6 -> 56.8Improved
add domain connectedness to camera invariance (ablation)Baseline + LC · Duke to Market-1501mAP: 28.5 -> 31.4Improved
add domain connectedness to camera invariance (ablation)Baseline + LC · Market-1501 to DukemAP: 22.1 -> 27.2Improved
add camera invariance loss (ablation)Baseline · Market-1501 to DukeRank-1 accuracy: 32.9 -> 42.5Improved
add camera invariance loss (ablation)Baseline · CUHK03 to DukeRank-1 accuracy: 24.3 -> 40.9Improved
add camera style transfer (vs crop or flip) as perturbationSource-only baseline · Duke to Market-1501Mean feature distance: 0.049 -> 0.485Mixed
use cluster-based or supervised instead of random sampling of target images (ablation)HHL · Duke/Market/CUHK03 transfersRank-1 accuracy and mAPNo clear effect
increase beta from 0 (baseline) to 0.4-0.8 (ablation)HHL · four transfer settingsRank-1 accuracy and mAPImproved

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