Fake-image detection with Robust Hashing

TL;DR

This paper explores the use of robust hashing techniques to detect manipulated fake images, including those altered by common compression methods, demonstrating superior performance over existing methods across various datasets.

Contribution

The paper introduces a novel fake-image detection method based on robust hashing that effectively identifies manipulated images even after multiple types of image alterations.

Findings

Outperforms state-of-the-art fake detection methods

Effective on images manipulated with JPEG compression and GANs

Demonstrates robustness across diverse datasets

Abstract

In this paper, we investigate whether robust hashing has a possibility to robustly detect fake-images even when multiple manipulation techniques such as JPEG compression are applied to images for the first time. In an experiment, the proposed fake detection with robust hashing is demonstrated to outperform state-of-the-art one under the use of various datasets including fake images generated with GANs.