An Improved Self-Training Model with Fine-Tuning Teacher/Student Exchange Strategy to Detect Computer-Generated Images

Table 12

Test results for different dataset-combination settings with attacks to 100% of the testing dataset.

Model	Accuracy (%)			Precision (%)			Recall (%)
	Comb1	Comb2	Comb3	Comb1	Comb2	Comb3	Comb1	Comb2	Comb3
M0	73.72	70.62	70.97	83.39	76.72	97.57	69.87	67.99	63.68
M1	80.98	79.82	88.06	91.02	83.43	94.82	75.78	77.80	83.52
M2	81.29	82.34	87.19	89.02	86.03	94.92	77.09	80.10	82.20
M3	81.35	80.75	88.16	89.57	81.40	94.75	76.91	80.34	83.71