Generalizable Person Re Identification On 21
Metriken
DukeMTMC-reID-u003eRank1
DukeMTMC-reID-u003emAP
MSMT17-u003eRank-1
MSMT17-u003emAP
MSMT17-All-u003eRank-1
MSMT17-All-u003emAP
RandPerson-u003eRank-1
RandPerson-u003emAP
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
| Modellname | DukeMTMC-reID-u003eRank1 | DukeMTMC-reID-u003emAP | MSMT17-u003eRank-1 | MSMT17-u003emAP | MSMT17-All-u003eRank-1 | MSMT17-All-u003emAP | RandPerson-u003eRank-1 | RandPerson-u003emAP |
|---|---|---|---|---|---|---|---|---|
| remix-training-generalized-person-re | 71.3 | 43.0 | 78.2 | 52.4 | 84.0 | 61.0 | 72.7 | 45.4 |
| graph-sampling-based-deep-metric-learning-for | - | - | 79.1 | 49.5 | 82.4 | 56.9 | 76.7 | 46.7 |
| transformer-based-deep-image-matching-for | - | - | 80.1 | 52.0 | 82.6 | 58.4 | 77.3 | 49.1 |
| interpretable-and-generalizable-deep-image | - | - | - | - | 72.6 | 43.1 | - | - |
| unleashing-the-potential-of-pre-trained | 71.5 | 42.3 | 77.9 | 52.1 | - | - | - | - |