Dual-Branch Convolutional Framework for Spatial and Frequency-Based Image Forgery Detection
Naman Tyagi, Riya Jain
TL;DR
The paper addresses the challenge of authenticating digital images in the era of deepfakes by proposing a dual-branch convolutional framework that fuses spatial and frequency-domain features. It adopts a Siamese network with contrastive learning to produce discriminative embeddings, achieving $64$-dimensional representations and demonstrating 77.9% accuracy on the CASIA 2.0 dataset. While not as high-performing as larger, more complex pipelines, the method balances detection reliability with computational efficiency, making it suitable for practical deployment in media verification, law, and forensics. Overall, the work advances visual forensics by integrating multi-domain features and deep metric learning to enhance robustness against sophisticated forgeries.
Abstract
With a very rapid increase in deepfakes and digital image forgeries, ensuring the authenticity of images is becoming increasingly challenging. This report introduces a forgery detection framework that combines spatial and frequency-based features for detecting forgeries. We propose a dual branch convolution neural network that operates on features extracted from spatial and frequency domains. Features from both branches are fused and compared within a Siamese network, yielding 64 dimensional embeddings for classification. When benchmarked on CASIA 2.0 dataset, our method achieves an accuracy of 77.9%, outperforming traditional statistical methods. Despite its relatively weaker performance compared to larger, more complex forgery detection pipelines, our approach balances computational complexity and detection reliability, making it ready for practical deployment. It provides a strong methodology for forensic scrutiny of digital images. In a broader sense, it advances the state of the art in visual forensics, addressing an urgent requirement in media verification, law enforcement and digital content reliability.