Table 4 The rank-1 accuracy and mAP with varying number of cropped regions on the Market-1501 dataset
From: Learning local embedding deep features for person re-identification in camera networks
K | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
Rank-1 | 75.03 | 77.65 | 80.77 | 76.32 | 73.86 |
mAP | 53.52 | 56.13 | 61.34 | 54.69 | 51.52 |