A Full-Image Full-Resolution End-to-End-Trainable CNN Framework for Image Forgery Detection

TL;DR

This paper introduces a full-resolution, end-to-end trainable CNN framework for image forgery detection that preserves high-frequency details by processing entire images without resizing, outperforming existing methods.

Contribution

It presents a novel CNN framework that enables full-image analysis at full resolution, overcoming memory limitations with gradient checkpointing and weak supervision.

Findings

Outperforms all baseline methods on forensic datasets.

Effectively preserves high-frequency details crucial for forgery detection.

Demonstrates strong generalization across multiple datasets.

Abstract

Due to limited computational and memory resources, current deep learning models accept only rather small images in input, calling for preliminary image resizing. This is not a problem for high-level vision problems, where discriminative features are barely affected by resizing. On the contrary, in image forensics, resizing tends to destroy precious high-frequency details, impacting heavily on performance. One can avoid resizing by means of patch-wise processing, at the cost of renouncing whole-image analysis. In this work, we propose a CNN-based image forgery detection framework which makes decisions based on full-resolution information gathered from the whole image. Thanks to gradient checkpointing, the framework is trainable end-to-end with limited memory resources and weak (image-level) supervision, allowing for the joint optimization of all parameters. Experiments on widespread image forensics datasets prove the good performance of the proposed approach, which largely outperforms all baselines and all reference methods.