EditScout: Locating Forged Regions from Diffusion-based Edited Images with Multimodal LLM
Quang Nguyen, Truong Vu, Trong-Tung Nguyen, Yuxin Wen, Preston K Robinette, Taylor T Johnson, Tom Goldstein, Anh Tran, Khoi Nguyen · 2024
Read EditScout: Locating Forged Regions from Diffusion-based Edited Images with Multimodal LLM by Quang Nguyen, Truong Vu, Trong-Tung Nguyen, Yuxin Wen, Preston K Robinette, Taylor T Johnson, Tom Goldstein, Anh Tran, Khoi Nguyen (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-03809.txt.
Description
Quang Nguyen and co-authors locate AI-edited regions in images in a 2024 paper. Modern editors can forge parts of a photo nearly invisibly to the eye. Classic forensics tools miss many diffusion-based edits entirely, leaving a gap.
Their framework pairs image analysis with a multimodal language model. The model helps name and outline the tampered region in the picture. Edited areas made by diffusion tools are the main target of the work. Media credibility and legal evidence both depend on such checks being reliable. Tests compare the system with existing forensic methods on hard cases. Readers see language models drafted into the fight against fake imagery.
Catalog record
| CREATOR | Quang Nguyen, Truong Vu, Trong-Tung Nguyen, Yuxin Wen, Preston K Robinette, Taylor T Johnson, Tom Goldstein, Anh Tran, Khoi Nguyen |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-03809