Table 16 Summary of the evaluation outcomes for the two suggested models.
From: Improved facial emotion recognition model based on a novel deep convolutional structure
Dataset | Proposed DCN (30 epochs) | Proposed AA-DCN (75 epochs) | ||||||
|---|---|---|---|---|---|---|---|---|
Train_acc (%) | Test_acc (%) | Avg epoch_ time(s) | Train- time (m) | Train_acc (%) | Test_acc (%) | Avg-epoch_ time(s) | Train - time (m) | |
CK+ | 98.09 | 98.32 | 6.64 | 3.32 | 98.89 | 99.26 | 4.2 | 5.25 |
JAFFE | 95.75 | 95 | 12 | 6 | 97.63 | 0.98 | 6.5 | 8.13 |
RAF_DB | 93.5 | 76 | 18 | 10 | 97 | 0.82 | 9.7 | 12.2 |
