Three Forensic Cues for JPEG AI Images

2025-04-04

AI Summary

JPEG AI—AI-based image compression—introduces forensic artifacts highly confounded with those of traditional JPEG and generative AI images (e.g., DeepFakes), rendering existing detection methods ineffective. To address this, we propose the first explainable digital forensics framework specifically designed for JPEG AI. Our method innovatively exploits three domain-specific cues: (i) cross-channel preprocessing correlations, modeled via chromatic channel statistics; (ii) degradation attenuation under repeated compression, analyzed through multi-generation distortion propagation; and (iii) latent-space quantization fingerprints, inferred via quantization step reverse engineering. These enable robust detection, re-compression identification, and authentic/synthetic image discrimination. Extensive experiments demonstrate consistently high accuracy—even at low bitrates—significantly outperforming state-of-the-art baselines. This work establishes the first interpretable, practically deployable foundation for judicial authentication of AI-compressed imagery.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Humans and AI: Other Foundations of Human Computation & AI
• Machine Learning: Learning on the Edge & Model Compression

Application Category

• Web Mining and Content Analysis: Web data provenance, reliability, and authenticity
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Search and Retrieval-Augmented AI: Agentic search

Abstract

The JPEG standard was vastly successful. Currently, the first AI-based compression method ``JPEG AI'' will be standardized. JPEG AI brings remarkable benefits. JPEG AI images exhibit impressive image quality at bitrates that are an order of magnitude lower than images compressed with traditional JPEG. However, forensic analysis of JPEG AI has to be completely re-thought: forensic tools for traditional JPEG do not transfer to JPEG AI, and artifacts from JPEG AI are easily confused with artifacts from artificially generated images (``DeepFakes''). This creates a need for novel forensic approaches to detection and distinction of JPEG AI images. In this work, we make a first step towards a forensic JPEG AI toolset. We propose three cues for forensic algorithms for JPEG AI. These algorithms address three forensic questions: first, we show that the JPEG AI preprocessing introduces correlations in the color channels that do not occur in uncompressed images. Second, we show that repeated compression of JPEG AI images leads to diminishing distortion differences. This can be used to detect recompression, in a spirit similar to some classic JPEG forensics methods. Third, we show that the quantization of JPEG AI images in the latent space can be used to distinguish real images with JPEG AI compression from synthetically generated images. The proposed methods are interpretable for a forensic analyst, and we hope that they inspire further research in the forensics of AI-compressed images.

Problem

• Detect JPEG AI image compression artifacts
• Identify recompression in JPEG AI images
• Distinguish JPEG AI from synthetic images

Innovation

• Detect color channel correlations in JPEG AI
• Identify recompression via diminishing distortion differences
• Distinguish real from synthetic using latent space quantization