Table 3 Parameters and model size comparison of models
From: Automatic segmentation of large-scale CT image datasets for detailed body composition analysis
Network | Trainable parameters (M) | Model size (MB) |
|---|---|---|
ResUNET | 0.81 | 3.3 |
UNET++ | 2.29 | 9.3 |
Ghost-UNET++ | 0.35 | 1.7 |