Dual JPEG Compatibility: a Reliable and Explainable Tool for Image Forensics

TL;DR

This paper introduces a compatibility-based method for JPEG image forensics that detects manipulations by analyzing whether image blocks are consistent with the original JPEG pipeline, showing promising results in controlled settings.

Contribution

It presents a novel approach to detect JPEG manipulations by finding block antecedents, outperforming some deep learning models in idealized scenarios.

Findings

Detects manipulations like inpainting, copy-move, and splicing after JPEG compression.

Recompression with higher quality factor reveals manipulations.

Guarantees zero false alarms with perfect pipeline knowledge.

Abstract

Given a JPEG pipeline (compression or decompression), this paper demonstrates how to find the antecedent of an 8x8 block. If it exists, the block is considered compatible with the pipeline. For unaltered images, all blocks remain compatible with the original pipeline; however, for manipulated images, this is not necessarily true. This article provides a first demonstration of the potential of compatibility-based approaches for JPEG image forensics. It introduces a method to address the key challenge of finding a block antecedent in a high-dimensional space, relying on a local search algorithm with restrictions on the search space. We show that inpainting, copy-move, and splicing, when applied after JPEG compression, result in three distinct mismatch problems that can be detected. In particular, if the image is re-compressed after modification, the manipulation can be detected when the quality factor of the second compression is higher than that of the first. Through extensive experiments, we highlight the potential of this compatibility attack under varying degrees of assumptions. While our approach shows promising results-outperforming three state-of-the-art deep learning models in an idealized setting-it remains a proof of concept rather than an off-the-shelf forensic tool. Notably, with a perfect knowledge of the JPEG pipeline, our method guarantees zero false alarms in block-by-block localization, given sufficient computational power.