Exposing Fake Images with Forensic Similarity Graphs

TL;DR

This paper introduces a novel graph-based approach for detecting and localizing image forgeries by modeling images as Forensic Similarity Graphs and applying community detection algorithms.

Contribution

The paper presents a new graph representation of images and adapts community detection techniques for improved forgery detection and localization.

Findings

Outperforms existing forgery detection methods

Effectively localizes tampered regions

Captures community structure in forged images

Abstract

We propose new image forgery detection and localization algorithms by recasting these problems as graph-based community detection problems. To do this, we introduce a novel abstract, graph-based representation of an image, which we call the Forensic Similarity Graph, that captures key forensic relationships among regions in the image. In this representation, small image patches are represented by graph vertices with edges assigned according to the forensic similarity between patches. Localized tampering introduces unique structure into this graph, which aligns with a concept called ``community structure'' in graph-theory literature. In the Forensic Similarity Graph, communities correspond to the tampered and unaltered regions in the image. As a result, forgery detection is performed by identifying whether multiple communities exist, and forgery localization is performed by partitioning these communities. We present two community detection techniques, adapted from literature, to detect and localize image forgeries. We experimentally show that our proposed community detection methods outperform existing state-of-the-art forgery detection and localization methods, which do not capture such community structure.