Papers › CORE-ReID: Comprehensive Optimization and Refinement through Ensemble Fusion in Domain...

CORE-ReID: Comprehensive Optimization and Refinement through Ensemble Fusion in Domain Adaptation for Person Re-Identification

3 Jun 2024Software 2024 6archive 2025-07-28

Trinh Quoc Nguyen, Oky Dicky Ardiansyah Prima, Katsuyoshi Hotta

This study introduces a novel framework, “Comprehensive Optimization and Refinement through Ensemble Fusion in Domain Adaptation for Person Re-identification (CORE-ReID)”, to address an Unsupervised Domain Adaptation (UDA) for Person Re-identification (ReID). The framework utilizes CycleGAN to generate diverse data that harmonize differences in image characteristics from different camera sources in the pre-training stage. In the fine-tuning stage, based on a pair of teacher–student networks, the framework integrates multi-view features for multi-level clustering to derive diverse pseudo-labels. A learnable Ensemble Fusion component that focuses on fine grained local information within global features is introduced to enhance learning comprehensiveness and avoid ambiguity associated with multiple pseudo-labels. Experimental results on three common UDAs in Person ReID demonstrated significant performance gains over state-of-the-art approaches. Additional enhancements, such as Efficient Channel Attention Block and Bidirectional Mean Feature Normalization mitigate deviation effects and the adaptive fusion of global and local features using the ResNet-based model, further strengthening the framework. The proposed framework ensures clarity in fusion features, avoids ambiguity, and achieves high accuracy in terms of Mean Average Precision, Top-1, Top-5, and Top-10, positioning it as an advanced and effective solution for UDA in Person ReID.

PaperPDFCode

Code

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Domain AdaptationPerson Re-IdentificationUnsupervised Domain AdaptationUnsupervised Person Re-Identification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Unsupervised Domain Adaptation CUHK03 to MSMT CORE-ReID R1 67.3 #2 of 7 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to MSMT CORE-ReID R10 83.1 #2 of 7 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to MSMT CORE-ReID R5 79.0 #2 of 7 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to MSMT CORE-ReID mAP 40.4 #2 of 7 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to Market CORE-ReID R1 93.6 #2 of 9 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to Market CORE-ReID R10 98.7 #2 of 9 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to Market CORE-ReID R5 97.3 #2 of 9 Archive leaderboard report
Unsupervised Domain Adaptation CUHK03 to Market CORE-ReID mAP 83.6 #2 of 9 Archive leaderboard report
Unsupervised Domain Adaptation Duke to MSMT CORE-ReID mAP 45.2 #1 of 13 Archive leaderboard report
Unsupervised Domain Adaptation Duke to MSMT CORE-ReID rank-1 72.2 #1 of 13 Archive leaderboard report
Unsupervised Domain Adaptation Duke to MSMT CORE-ReID rank-10 86.3 #1 of 13 Archive leaderboard report
Unsupervised Domain Adaptation Duke to MSMT CORE-ReID rank-5 82.9 #1 of 13 Archive leaderboard report
Unsupervised Domain Adaptation Duke to Market CORE-ReID mAP 84.4 #1 of 26 Archive leaderboard report
Unsupervised Domain Adaptation Duke to Market CORE-ReID rank-1 93.6 #1 of 26 Archive leaderboard report
Unsupervised Domain Adaptation Duke to Market CORE-ReID rank-10 98.7 #1 of 26 Archive leaderboard report
Unsupervised Domain Adaptation Duke to Market CORE-ReID rank-5 97.7 #1 of 26 Archive leaderboard report
Unsupervised Domain Adaptation Market to CUHK03 CORE-ReID R1 61.0 #2 of 8 Archive leaderboard report
Unsupervised Domain Adaptation Market to CUHK03 CORE-ReID R10 87.2 #2 of 8 Archive leaderboard report
Unsupervised Domain Adaptation Market to CUHK03 CORE-ReID R5 79.6 #2 of 8 Archive leaderboard report
Unsupervised Domain Adaptation Market to CUHK03 CORE-ReID mAP 62.9 #2 of 8 Archive leaderboard report
Unsupervised Domain Adaptation Market to Duke CORE-ReID mAP 74.8 #1 of 25 Archive leaderboard report
Unsupervised Domain Adaptation Market to Duke CORE-ReID rank-1 84.8 #1 of 25 Archive leaderboard report
Unsupervised Domain Adaptation Market to Duke CORE-ReID rank-10 94.4 #1 of 25 Archive leaderboard report
Unsupervised Domain Adaptation Market to Duke CORE-ReID rank-5 92.4 #1 of 25 Archive leaderboard report
Unsupervised Domain Adaptation Market to MSMT CORE-ReID mAP 41.9 #2 of 17 Archive leaderboard report
Unsupervised Domain Adaptation Market to MSMT CORE-ReID rank-1 69.5 #2 of 17 Archive leaderboard report
Unsupervised Domain Adaptation Market to MSMT CORE-ReID rank-10 84.4 #2 of 17 Archive leaderboard report
Unsupervised Domain Adaptation Market to MSMT CORE-ReID rank-5 80.3 #2 of 17 Archive leaderboard report
Unsupervised Person Re-Identification DukeMTMC-reID->MSMT17 CORE-ReID Rank-1 72.2 #1 of 7 Archive leaderboard report
Unsupervised Person Re-Identification DukeMTMC-reID->MSMT17 CORE-ReID Rank-10 86.3 #1 of 7 Archive leaderboard report
Unsupervised Person Re-Identification DukeMTMC-reID->MSMT17 CORE-ReID Rank-5 82.9 #1 of 7 Archive leaderboard report
Unsupervised Person Re-Identification DukeMTMC-reID->MSMT17 CORE-ReID mAP 45.2 #1 of 7 Archive leaderboard report
Unsupervised Person Re-Identification DukeMTMC-reID->Market-1501 CORE-ReID Rank-1 93.6 #1 of 8 Archive leaderboard report
Unsupervised Person Re-Identification DukeMTMC-reID->Market-1501 CORE-ReID Rank-10 98.7 #1 of 8 Archive leaderboard report
Unsupervised Person Re-Identification DukeMTMC-reID->Market-1501 CORE-ReID Rank-5 97.7 #1 of 8 Archive leaderboard report
Unsupervised Person Re-Identification DukeMTMC-reID->Market-1501 CORE-ReID mAP 84.4 #1 of 8 Archive leaderboard report
Unsupervised Person Re-Identification Market-1501->DukeMTMC-reID CORE-ReID Rank-1 84.8 #1 of 7 Archive leaderboard report
Unsupervised Person Re-Identification Market-1501->DukeMTMC-reID CORE-ReID Rank-10 94.4 #1 of 7 Archive leaderboard report
Unsupervised Person Re-Identification Market-1501->DukeMTMC-reID CORE-ReID Rank-5 92.4 #1 of 7 Archive leaderboard report
Unsupervised Person Re-Identification Market-1501->DukeMTMC-reID CORE-ReID mAP 74.8 #1 of 7 Archive leaderboard report
Unsupervised Person Re-Identification Market-1501->MSMT17 CORE-ReID Rank-1 69.5 #1 of 7 Archive leaderboard report
Unsupervised Person Re-Identification Market-1501->MSMT17 CORE-ReID Rank-10 84.4 #1 of 7 Archive leaderboard report
Unsupervised Person Re-Identification Market-1501->MSMT17 CORE-ReID Rank-5 80.3 #1 of 7 Archive leaderboard report
Unsupervised Person Re-Identification Market-1501->MSMT17 CORE-ReID mAP 41.9 #1 of 7 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

1x1 ConvolutionAverage PoolingBatch NormalizationConvolutionCycle Consistency LossEfficient Channel AttentionGAN Least Squares LossGlobal Average PoolingInstance NormalizationPatchGANReLUResidual BlockResidual ConnectionSigmoid ActivationTanh Activation

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections