Table 3 Computational efficiency comparison of SPAYOLO and baseline models in terms of FLOPs, parameter count, training time per epoch, and average inference time per image.
From: Spectral-spatial feature fusion for real-time facial expression recognition
Model | Computation (GFLOPs) | Parameters (M) | Training Time per Epoch (s) | Inference Time per Image (ms) |
|---|---|---|---|---|
SPAYOLO(ours) | 14.4 | 5.7 | 13 | 2.1 |
ResEmoteNet | 4.35 | 80.24 | 43 | 5.8 |
EmoNeXt-Tiny-22k | 4.57 | 30.56 | 150 | 8.4 |
