Table 4 Performance comparison on defacto (splicing) dataset.

From: Multi-resolution transfer learning for tampered image classification using SE-enhanced fused-MBConv and optimized CNN heads

Dataset: defacto (splicing)

Ref. No.

Model

Year

AUC

F1 score

Accuracy

Precision

Recall

25

CAT-Net

2023

–

–

0.99

–

–

25

DRRU-Net

2023

–

–

0.99

–

–

25

DRRU-Net LR

2023

–

–

0.99

–

–

25

DRRU-Net TL

2023

–

–

0.99

–

–

25

RRU-Net

2023

–

–

0.988

–

–

Proposed

CNN+EfficientNetV2B0

2025

1.0

0.9997

0.9997

0.9997

0.9997

Back to article page