Table 1 For dataset evaluation, we use the ResNet-50 with triplet loss function and the proposed model with a combination of different loss functions
Network (dataset) | Margin 0.1 | Margin 0.2 | Margin 0.3 | |||
|---|---|---|---|---|---|---|
| Â | mAp (%) | Rank-1 (%) | mAp (%) | Rank-1 (%) | mAp (%) | Rank-1 (%) |
ResNet-50 (ImageNet) | 53.0 | 62.1 | 57.8 | 65.4 | 62.6 | 65.0 |
ResNet-50 (cuhk-sysu) | 70.8 | 72.5 | 72.7 | 75.5 | 70.5 | 72.5 |
ResNet-50 (Market1501) | 69.7 | 70.4 | 70.7 | 70.6 | 69.7 | 72.1 |
ResNet-50 (Cuhk03) | 71.7 | 73.6 | 70.5 | 71.3 | 68.8 | 71.7 |
triplet loss + L-GM loss | ||||||
ResNet-50 (ImageNet) | 54.3 | 64.2 | 59.1 | 67.3 | 61.4 | 64.2 |
ResNet-50 (cuhk-sysu) | 68.1 | 71.9 | 68.2 | 73.2 | 69.6 | 71.1 |
ResNet-50 (Market1501) | 67.2 | 70.7 | 67.1 | 72.7 | 61.7 | 68.7 |
ResNet-50 (Cuhk03) | 63.2 | 67.0 | 67.1 | 72.4 | 68.8 | 71.7 |