Addressing Image Authenticity When Cameras Use Generative AI

2026-04-23

AI Summary

This work addresses the challenge of semantic “hallucinations” inadvertently introduced by generative AI integrated into modern image signal processing (ISP) pipelines, which compromise photographic authenticity in ways often imperceptible to users. To counteract this without requiring access to the original ISP, the authors propose a post-hoc restoration method that jointly optimizes an image-specific multilayer perceptron (MLP) decoder and a modality encoder to reconstruct faithful, unaltered content from processed images. Innovatively, they embed an ultra-lightweight model—merely 180 KB in size—within standard image metadata (e.g., JPEG or HEIC), enabling, for the first time, reversible removal of hallucinations induced by AI-based enhancements such as digital zoom or low-light processing, thereby restoring semantically accurate and trustworthy imagery.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Multi-modal Vision
• Natural Language Processing: Generation

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

The ability of generative AI (GenAI) methods to photorealistically alter camera images has raised awareness about the authenticity of images shared online. Interestingly, images captured directly by our cameras are considered authentic and faithful. However, with the increasing integration of deep-learning modules into cameras' capture-time hardware -- namely, the image signal processor (ISP) -- there is now a potential for hallucinated content in images directly output by our cameras. Hallucinated capture-time image content is typically benign, such as enhanced edges or texture, but in certain operations, such as AI-based digital zoom or low-light image enhancement, hallucinations can potentially alter the semantics and interpretation of the image content. As a result, users may not realize that the content in their camera images is not authentic. This paper addresses this issue by enabling users to recover the 'unhallucinated' version of the camera image to avoid misinterpretation of the image content. Our approach works by optimizing an image-specific multi-layer perceptron (MLP) decoder together with a modality-specific encoder so that, given the camera image, we can recover the image before hallucinated content was added. The encoder and MLP are self-contained and can be applied post-capture to the image without requiring access to the camera ISP. Moreover, the encoder and MLP decoder require only 180 KB of storage and can be readily saved as metadata within standard image formats such as JPEG and HEIC.

Problem

• Image Authenticity
• Generative AI
• Hallucinated Content
• Camera ISP
• Digital Image Integrity

Innovation

• generative AI
• image authenticity
• hallucination removal
• image signal processor
• MLP decoder