EditTrack: Detecting and Attributing AI-assisted Image Editing

2025-10-01

AI Summary

This work formally defines and addresses, for the first time, the detection and attribution problem in AI-assisted image editing: given a source image and a suspect image, determine whether the latter was generated from the former via a specific AI editing model and identify that model. To this end, we propose the first unified framework for editing provenance, featuring three core innovations: (1) a re-editing strategy to synthesize discriminative editing traces; (2) a fine-grained similarity metric based on deep features; and (3) joint modeling of editing trajectories with multi-model contrastive learning. Extensive experiments across five state-of-the-art AI editing models—including InstructPix2Pix and DragGAN—and six benchmark datasets demonstrate that our method significantly outperforms five SOTA baselines on both detection and attribution tasks, achieving an average accuracy improvement of 12.7%. Moreover, it exhibits strong generalization across unseen models and datasets.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Web Mining and Content Analysis: Web data provenance, reliability, and authenticity
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Security and Privacy: Data transparency and provenance

Abstract

In this work, we formulate and study the problem of image-editing detection and attribution: given a base image and a suspicious image, detection seeks to determine whether the suspicious image was derived from the base image using an AI editing model, while attribution further identifies the specific editing model responsible. Existing methods for detecting and attributing AI-generated images are insufficient for this problem, as they focus on determining whether an image was AI-generated/edited rather than whether it was edited from a particular base image. To bridge this gap, we propose EditTrack, the first framework for this image-editing detection and attribution problem. Building on four key observations about the editing process, EditTrack introduces a novel re-editing strategy and leverages carefully designed similarity metrics to determine whether a suspicious image originates from a base image and, if so, by which model. We evaluate EditTrack on five state-of-the-art editing models across six datasets, demonstrating that it consistently achieves accurate detection and attribution, significantly outperforming five baselines.

Problem

• Detecting AI-edited images derived from specific base images
• Identifying the exact AI model used for image editing
• Distinguishing edited images from original ones accurately

Innovation

• Re-editing strategy for image editing detection
• Similarity metrics for model attribution
• Framework for detecting AI-edited images from base