AI Paper Atlas

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

Self-Similarity Grouping: A Simple Unsupervised Cross Domain Adaptation Approach for Person Re-Identification

Yang Fu, Yunchao Wei, Guanshuo Wang, Yuqian Zhou, Honghui Shi, Thomas S. Huang

ICCV 2019 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Geometric alignment and synthetic-to-real adaptation · 573 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 Self-similarity Grouping (global + upper + lower part pseudo labels) to the source-trained modelBaseline · DukeMTMC-reID -> Market1501mAP: 26.6 -> 58.3Improved
add Self-similarity Grouping to the source-trained modelBaseline · Market1501 -> DukeMTMC-reIDmAP: 16.1 -> 53.4Improved
replace UDA pseudo-label method with SSGBaseline · Market1501 -> DukeMTMC-reIDmAP: 49.0 -> 53.4Improved
split features into 2 parts instead of 3 for SSG (ablation)SSG · DukeMTMC-reID -> Market1501mAP: 50.4 -> 53.4Improved
split features into 2 parts instead of 4 for SSG (ablation)SSG · DukeMTMC-reID -> Market1501mAP: 49.1 -> 53.4Improved
add clustering-guided semi-supervised training to SSG (ablation)SSG · DukeMTMC-reID -> Market1501mAP: 58.3 -> 62.5Improved
add joint training of SSG and semi-supervised losses instead of fine-tuning (ablation)SSG+ · DukeMTMC-reID -> Market1501mAP: 62.5 -> 68.7Improved
add joint training of SSG and semi-supervised losses instead of fine-tuning (ablation)SSG+ · Market1501 -> DukeMTMC-reIDmAP: 56.7 -> 60.3Improved
use clustering-guided annotation instead of random sampling annotation (ablation)SSG++ · Market1501 -> DukeMTMC-reIDmAP: 57.6 -> 60.3Improved
use clustering-guided annotation instead of random sampling annotation (ablation)SSG++ · DukeMTMC-reID -> Market1501mAP: 66.8 -> 68.7Improved
add semi-supervised joint training to SSG on MSMT17 targetSSG · DukeMTMC-reID -> MSMT17mAP: 13.3 -> 18.3Improved

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