FlipReID (without re-ranking)

Reported on 6 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 Vision6 results

Person Re-IdentificationonMSMT17
Rank-1· 2021-05-12
85.6
best: 91.7 (SOLIDER (with re-ranking))
FlipReID: Closing the Gap between Training and Inference in Person Re-Identification arXiv:2105.05639
Person Re-IdentificationonMSMT17
mAP· 2021-05-12
68
best: 86.7 (CLIP-ReID (with re-ranking))
FlipReID: Closing the Gap between Training and Inference in Person Re-Identification arXiv:2105.05639
Person Re-IdentificationonMarket-1501
Rank-1· 2021-05-12
95.5
best: 98 (st-ReID(RE, RK))
FlipReID: Closing the Gap between Training and Inference in Person Re-Identification arXiv:2105.05639
Person Re-IdentificationonMarket-1501
mAP· 2021-05-12
89.6
best: 96.21 (Unsupervised Pre-training (ResNet101+RK))
FlipReID: Closing the Gap between Training and Inference in Person Re-Identification arXiv:2105.05639
Person Re-IdentificationonDukeMTMC-reID
Rank-1· 2021-05-12
90.9
best: 95.6 (CTL Model (ResNet50, 256x128))
FlipReID: Closing the Gap between Training and Inference in Person Re-Identification arXiv:2105.05639
Person Re-IdentificationonDukeMTMC-reID
mAP· 2021-05-12
81.5
best: 97.1 (DenseIL)
FlipReID: Closing the Gap between Training and Inference in Person Re-Identification arXiv:2105.05639