Table 1 Performance Evaluation of Deep neural network architectures
From: “Idol talks!” AI-driven image to text to speech: illustrated by an application to images of deities
Deep neural network model | Accuracy | Precision | Recall | F1-Score |
|---|---|---|---|---|
DenseNet 121 | 74.90 | 0.74 | 0.68 | 0.72 |
Resnet 50 | 63.52 | 0.43 | 0.53 | 0.63 |
EfficientDet | 64.23 | 0.59 | 0.53 | 0.61 |
Cascade R-CNN | 76.82 | 0.69 | 0.76 | 0.77 |
FAST R-CNN | 80.73 | 0.79 | 0.76 | 0.74 |
YOLO V3 | 83.50 | 0.83 | 0.72 | 0.86 |
YOLO V5 | 96.75 | 0.86 | 0.90 | 0.91 |
