Saved in:
Bibliographic Details
Main Authors: Costanzino, Alex, Bayliss, Woody, Sock, Juil, Blanch, Marc Gorriz, Horak, Danijela, Laptev, Ivan, Torr, Philip, Pizzati, Fabio
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.05466
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909646595293184
author Costanzino, Alex
Bayliss, Woody
Sock, Juil
Blanch, Marc Gorriz
Horak, Danijela
Laptev, Ivan
Torr, Philip
Pizzati, Fabio
author_facet Costanzino, Alex
Bayliss, Woody
Sock, Juil
Blanch, Marc Gorriz
Horak, Danijela
Laptev, Ivan
Torr, Philip
Pizzati, Fabio
contents Changing facial expressions, gestures, or background details may dramatically alter the meaning conveyed by an image. Notably, recent advances in diffusion models greatly improve the quality of image manipulation while also opening the door to misuse. Identifying changes made to authentic images, thus, becomes an important task, constantly challenged by new diffusion-based editing tools. To this end, we propose a novel approach for ReliAble iDentification of inpainted AReas (RADAR). RADAR builds on existing foundation models and combines features from different image modalities. It also incorporates an auxiliary contrastive loss that helps to isolate manipulated image patches. We demonstrate these techniques to significantly improve both the accuracy of our method and its generalisation to a large number of diffusion models. To support realistic evaluation, we further introduce BBC-PAIR, a new comprehensive benchmark, with images tampered by 28 diffusion models. Our experiments show that RADAR achieves excellent results, outperforming the state-of-the-art in detecting and localising image edits made by both seen and unseen diffusion models. Our code, data and models will be publicly available at https://alex-costanzino.github.io/radar/.
format Preprint
id arxiv_https___arxiv_org_abs_2506_05466
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Reliable Identification of Diffusion-based Image Manipulations
Costanzino, Alex
Bayliss, Woody
Sock, Juil
Blanch, Marc Gorriz
Horak, Danijela
Laptev, Ivan
Torr, Philip
Pizzati, Fabio
Computer Vision and Pattern Recognition
Changing facial expressions, gestures, or background details may dramatically alter the meaning conveyed by an image. Notably, recent advances in diffusion models greatly improve the quality of image manipulation while also opening the door to misuse. Identifying changes made to authentic images, thus, becomes an important task, constantly challenged by new diffusion-based editing tools. To this end, we propose a novel approach for ReliAble iDentification of inpainted AReas (RADAR). RADAR builds on existing foundation models and combines features from different image modalities. It also incorporates an auxiliary contrastive loss that helps to isolate manipulated image patches. We demonstrate these techniques to significantly improve both the accuracy of our method and its generalisation to a large number of diffusion models. To support realistic evaluation, we further introduce BBC-PAIR, a new comprehensive benchmark, with images tampered by 28 diffusion models. Our experiments show that RADAR achieves excellent results, outperforming the state-of-the-art in detecting and localising image edits made by both seen and unseen diffusion models. Our code, data and models will be publicly available at https://alex-costanzino.github.io/radar/.
title Towards Reliable Identification of Diffusion-based Image Manipulations
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.05466