Table 4 Comparison of efficiency with state-of-the-art methods on the datasets of PET2009 and MS COCO.
Model | Datasets | |||||||
|---|---|---|---|---|---|---|---|---|
PET2009 | MS COCO | |||||||
Parameters(M) | Flops(G) | Inference Time(ms) | Trainning Time(s) | Parameters(M) | Flops(G) | Inference Time(ms) | Trainning Time(s) | |
KFC29 | 547.46 | 5.04 | 8.39 | 505.53 | 551.13 | 6.37 | 8.30 | 566.82 |
Shay et al.30 | 783.82 | 8.29 | 11.24 | 784.42 | 772.67 | 8.34 | 11.81 | 700.07 |
Yong et al.31 | 425.39 | 5.96 | 12.69 | 655.42 | 737.92 | 7.63 | 7.56 | 427.82 |
Fang et al.32 | 664.46 | 7.92 | 11.57 | 780.17 | 745.57 | 8.67 | 11.13 | 688.23 |
HDT network33 | 462.92 | 4.45 | 7.81 | 397.11 | 394.94 | 4.80 | 8.34 | 490.38 |
L-YOLOv434 | 338.15 | 3.52 | 5.33 | 326.57 | 318.76 | 3.65 | 5.65 | 338.20 |
Ours | 331.28 | 3.21 | 5.05 | 324.08 | 317.25 | 3.43 | 5.62 | 336.06 |
