OSNet

Reported on 5 benchmarks across 1 task · 1 paper · 2 SOTA

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

Computer Vision5 results

Person Re-IdentificationonMSMT17
Rank-1· 2019-05-02
78.7
best: 91.7 (SOLIDER (with re-ranking))
SOTA
Omni-Scale Feature Learning for Person Re-Identification arXiv:1905.00953
Person Re-IdentificationonMSMT17
mAP· 2019-05-02
52.9
best: 86.7 (CLIP-ReID (with re-ranking))
SOTA
Omni-Scale Feature Learning for Person Re-Identification arXiv:1905.00953
Person Re-IdentificationonMarket-1501
Rank-1· uses extra data· 2019-05-02
94.8
best: 98 (st-ReID(RE, RK))
Omni-Scale Feature Learning for Person Re-Identification arXiv:1905.00953
Person Re-IdentificationonMarket-1501
mAP· uses extra data· 2019-05-02
84.9
best: 96.21 (Unsupervised Pre-training (ResNet101+RK))
Omni-Scale Feature Learning for Person Re-Identification arXiv:1905.00953
Person Re-IdentificationonCUHK03
MAP· 2019-05-02
67.8
best: 94.3 (Proposed SGGNN)
Omni-Scale Feature Learning for Person Re-Identification arXiv:1905.00953