Table 5 Quantitative comparison of human re-id across cameras on a quite challenging VIPeR benchmark dataset
Item no. | Used human re-id method | Rank on VIPeR | Reference | ||
|---|---|---|---|---|---|
1 | 10 | 20 | |||
1 | ELF | 12.00 | 44.00 | 61.00 | 2008 ECCV [39] |
2 | SDALF | 19.87 | 49.37 | 65.73 | 2010 CVPR [40] |
3 | ColorInv | 24.21 | 57.09 | 69.65 | 2013 TPAMI [41] |
4 | SCNCD | 37.80 | 81.20 | 90.40 | 2014 ECCV [42] |
5 | LOMO + XQDA | 40.00 | 80.51 | 91.08 | 2015 CVPR [43] |
6 | FFN | 51.1 | 91.4 | 96.9 | 2016 WACV [44] |
7 | KISSME | 19.6 | 62.2 | 77.0 | 2012CVPR [45] |
8 | LFDA | 24.18 | 67.12 | 82.00 | 2013 CVPR [46] |
9 | KLFDA | 32.3 | 79.7 | 90.9 | 2014 ECCV [47] |
10 | MetricEnsb | 45.9 | 88.9 | 95.8 | 2015 CVPR [48] |
11 | LSSL | 47.8 | 87.6 | 94.2 | 2016 AAAI [49] |
12 | SCSP | 53.5 | 91.5 | 96.6 | 2016 CVPR [50] |