JPEGs Just Got Snipped: Croppable Signatures Against Deepfake Images

View on arXiv ← Back to list

Authors: Pericle Perazzo, Massimiliano Mattei, Giuseppe Anastasi, Marco Avvenuti, Gianluca Dini, Giuseppe Lettieri, Carlo Vallati

Published: 2025-12-01 16:30:53+00:00

AI Summary

This paper proposes a novel cryptographic method for image authentication that leverages the aggregability of Boneh, Lynn, and Shacham (BLS) signatures. The key contribution is implementing signatures that remain valid after legitimate image cropping, while simultaneously being invalidated by all other manipulations, including deepfake creation. By achieving an O(1) signature size for the cropped image, the solution offers high bandwidth efficiency, making it practical for web server dissemination scenarios.

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.

Key findings

The BLS-based scheme produces signed full and cropped images that are substantially smaller than a similar construction using standard, non-aggregable signatures (Johnson et al.). This size advantage is particularly noticeable at finer block granularities (16x16 pixels or smaller). The O(1) size of the cropped signature minimizes bandwidth consumption, proving the system's practicality for media distribution where cropping is a frequent transformation.

Approach

The proposed method divides the image into pixel blocks (e.g., 8x8 in JPEG) and uses an ephemeral private key to generate a BLS signature for each block. When the image is legitimately cropped, the cropper aggregates the signatures of the selected blocks to produce a constant-sized O(1) cropped signature, which is then verified by the client using the signer's public key.

Datasets

UNKNOWN

Model(s)

BLS signature scheme (BN-158), ECDSA (secp160r1)

Author countries

Italy

← Previous