Table 4 Comparison of inference speed for different models.
From: Character generation and visual quality enhancement in animated films using deep learning
Model | Inference time (ms) | GPU memory consumption (MB) | Frames per second (FPS) |
|---|---|---|---|
E-FOMM | 100 | 2550 | 10.00 |
FOMM | 85 | 2300 | 11.76 |
FOMM + CBAM | 92 | 2400 | 10.87 |
FOMM + Transformer | 250 | 3500 | 4.00 |
