JPEGs Just Got Snipped: Croppable Signatures Against Deepfake Images

TL;DR

This paper introduces a novel cryptographic signature scheme based on BLS signatures that remains valid after image cropping, specifically designed to detect deepfake manipulations in JPEG images, enhancing media authenticity verification.

Contribution

It proposes a crop-robust signature method using BLS signatures adapted for JPEG images, which is practical, size-efficient, and does not require trust in the image cropper.

Findings

Signature remains valid after cropping but invalidates other manipulations.

The scheme is O(1) in signature size, suitable for web dissemination.

Experimental results show practical signed JPEG image sizes.

Abstract

Deepfakes are a type of synthetic media created using artificial intelligence, specifically deep learning algorithms. This technology can for example superimpose faces and voices onto videos, creating hyper-realistic but artificial representations. Deepfakes pose significant risks regarding misinformation and fake news, because they can spread false information by depicting public figures saying or doing things they never did, undermining public trust. In this paper, we propose a method that leverages BLS signatures (Boneh, Lynn, and Shacham 2004) to implement signatures that remain valid after image cropping, but are invalidated in all the other types of manipulation, including deepfake creation. Our approach does not require who crops the image to know the signature private key or to be trusted in general, and it is O(1) in terms of signature size, making it a practical solution for scenarios where images are disseminated through web servers and cropping is the primary transformation. Finally, we adapted the signature scheme for the JPEG standard, and we experimentally tested the size of a signed image.