Unsupervised Person Re Identification On 5
Metriken
MAP
Rank-1
Rank-10
Rank-5
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
| Modellname | MAP | Rank-1 | Rank-10 | Rank-5 |
|---|---|---|---|---|
| ice-inter-instance-contrastive-encoding-for | 69.9 | 83.3 | 94.1 | 91.5 |
| one-shot-domain-adaptation-for-person-re | 55.9 | 72.4 | 87.7 | 84 |
| self-paced-contrastive-learning-with-hybrid | 65.3 | 81.2 | 92.2 | 90.3 |
| invariance-matters-exemplar-memory-for-domain | 40.4 | 63.3 | 80.4 | 75.8 |
| learning-transferable-pedestrian | 74.9 | 86.1 | - | - |
| person-transfer-gan-to-bridge-domain-gap-for | - | 27.4 | 50.7 | - |
| image-image-domain-adaptation-with-preserved | 26.2 | 46.4 | 68.0 | 62.3 |
| transformer-based-multi-grained-features-for | 76.8 | 86.7 | 94.1 | 92.9 |
| group-sampling-for-unsupervised-person-re | 69.1 | 82.7 | 93.5 | 91.1 |
| cluster-guided-asymmetric-contrastive | 69.6 | 82.6 | 93.8 | 91.2 |
| unsupervised-person-re-identification | 16.4 | 30.0 | 48.5 | 43.4 |
| camera-aware-proxies-for-unsupervised-person | 67.3 | 81.1 | 91.8 | 89.3 |
| hard-sample-guided-hybrid-contrast-learning | 73.3 | 85.1 | 94.6 | 92.4 |