AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

Unlabeled Samples Generated by GAN Improve the Person Re-Identification Baseline in Vitro

Zhedong Zheng, Liang Zheng, Yi Yang

ICCV 2017 · Label-efficient and robust learning › Semi-supervised learning · approach: GAN-based semi-supervised learning · 2,143 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 24,000 GAN images with LSROResNet-50 baseline · Market-1501rank-1 accuracy (%): 73.69 -> 78.06Improved
add 24,000 GAN images with LSROResNet-50 baseline · Market-1501mAP (%): 51.48 -> 56.23Improved
add GAN images with LSROResNet-50 baseline · DukeMTMC-reIDrank-1 accuracy (%): 65.22 -> 67.68Improved
add GAN images (from Market-1501) with LSROResNet-50 baseline · CUHK03rank-1 accuracy (%): 71.5 -> 73.1Improved
add GAN images with LSROVerif-Identif. model · Market-1501rank-1 accuracy (%): 79.51 -> 83.97Improved
add GAN images with LSROResNet-50 baseline · CUB-200-2011top-1 accuracy (%): 82.6 -> 83.2Improved
increase GAN images from 24,000 to 36,000 (ablation)LSRO · Market-1501rank-1 accuracy (%): 78.06 -> 76.07Hurt
use 12,000 instead of 24,000 GAN images (ablation)LSRO · Market-1501rank-1 accuracy (%): 78.06 -> 76.81Hurt
use LSRO instead of All in one (ablation)GAN-augmented training · Market-1501 (24,000 GAN images)rank-1 accuracy (%): 76.63 -> 78.06Improved
use LSRO instead of Pseudo label (ablation)GAN-augmented training · Market-1501 (24,000 GAN images)rank-1 accuracy (%): 75.80 -> 78.06Improved
use GAN images instead of real CUHK03 images as unlabeled data (ablation)LSRO · Market-1501 (12,000 unlabeled images)rank-1 accuracy (%): 75.65 -> 76.81Improved
add real CUHK03 images as unlabeled data (ablation)LSRO · Market-1501rank-1 accuracy (%): 73.69 -> 75.65Improved

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