Pyramid (CVPR'19)

Reported on 6 benchmarks across 1 task · 1 paper · 3 SOTA

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

Computer Vision6 results

Person Re-IdentificationonCUHK03 detected
MAP· 2018-10-29
74.8
best: 86.9 (Top-DB-Net + RK)
SOTA
Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training arXiv:1810.12193
Person Re-IdentificationonCUHK03 detected
Rank-1· 2018-10-29
78.9
best: 85.7 (Top-DB-Net + RK)
SOTA
Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training arXiv:1810.12193
Person Re-IdentificationonDukeMTMC-reID
Rank-1· 2018-10-29
89
best: 95.6 (CTL Model (ResNet50, 256x128))
SOTA
Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training arXiv:1810.12193
Person Re-IdentificationonMarket-1501
Rank-1· 2018-10-29
95.7
best: 98 (st-ReID(RE, RK))
Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training arXiv:1810.12193
Person Re-IdentificationonMarket-1501
mAP· 2018-10-29
88.2
best: 96.21 (Unsupervised Pre-training (ResNet101+RK))
Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training arXiv:1810.12193
Person Re-IdentificationonDukeMTMC-reID
mAP· 2018-10-29
79
best: 97.1 (DenseIL)
Pyramidal Person Re-IDentification via Multi-Loss Dynamic Training arXiv:1810.12193