Table 1 Accuracy comparison with state-of-the-art methods on the datasets of MOT2015 and MOT2016.
Model | Datasets | |||||||
|---|---|---|---|---|---|---|---|---|
MOT2015 | MOT2016 | |||||||
Accuracy(%) | Recall(%) | F1 Sorce(%) | AUC(%) | Accuracy(%) | Recall(%) | F1 Sorce(%) | AUC(%) | |
KFC29 | 91.7 | 85.69 | 84.45 | 92.25 | 87.57 | 90.23 | 87.11 | 92.71 |
Shay et al.30 | 86.38 | 89.96 | 86.28 | 87.87 | 86.91 | 89.99 | 89.05 | 88.67 |
Yong et al.31 | 87.5 | 89.81 | 86.42 | 91.28 | 92.56 | 85.49 | 88.22 | 93.44 |
Fang et al.32 | 88.08 | 89.46 | 86.07 | 88.08 | 90.32 | 84.71 | 85.02 | 88.12 |
HDT network33 | 91.69 | 88.72 | 87.4 | 85.64 | 92.11 | 90.24 | 84.98 | 89.58 |
L-YOLOv434 | 92.07 | 92.29 | 90.9 | 88.27 | 92.79 | 87.7 | 85.39 | 90.25 |
Ours | 93.22 | 95.63 | 92.18 | 92.38 | 95.58 | 92.43 | 93.25 | 96.22 |
