Research Article
Designing Compact Convolutional Filters for Lightweight Human Pose Estimation
Table 4
Comparisons on the MPII val set. The GFLOPs is computed with the input size
.
| Model | #Params | GFLOPs | PCKh |
| MobileNetV2 1× | 9.6M | 1.97 | 85.4 | MobileNetV3 1× | 8.7M | 1.82 | 84.3 | ShuffleNetV2 1× | 7.6M | 1.70 | 82.8 | Small HRNet-W16 | 1.3M | 0.72 | 80.2 | Lite-HRNet-18 | 1.1M | 0.27 | 86.1 | MobilePoseNet | 1.5M | 0.74 | 87.3 |
|
|