Cluster-level Alignment (Resnet50 w/o RK)

Reported on 4 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision4 results

Person Re-IdentificationonMarket-1501
Rank-1· 2020-08-15
95.7
best: 98 (st-ReID(RE, RK))
Cluster-level Feature Alignment for Person Re-identification arXiv:2008.06810
Person Re-IdentificationonMarket-1501
mAP· 2020-08-15
89.5
best: 96.21 (Unsupervised Pre-training (ResNet101+RK))
Cluster-level Feature Alignment for Person Re-identification arXiv:2008.06810
Person Re-IdentificationonDukeMTMC-reID
Rank-1· 2020-08-15
91.11
best: 95.6 (CTL Model (ResNet50, 256x128))
Cluster-level Feature Alignment for Person Re-identification arXiv:2008.06810
Person Re-IdentificationonDukeMTMC-reID
mAP· 2020-08-15
81.84
best: 97.1 (DenseIL)
Cluster-level Feature Alignment for Person Re-identification arXiv:2008.06810