.png){kind=icon}
Return
A modular image manipulation localization framework using a dual-stream classifier and conditional diffusion models
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-026-21940-0.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Image manipulation localization (IML), the task of accurately identifying the manipulated regions within an image, remains a significant challenge. Conventional deep learning methods often struggle with generalization and adapting to new manipulation schemes. To address these limitations, a novel two-stage IML system is proposed. The first stage employs a Dual-Stream Manipulation Classifier, fusing features from the standard RGB domain with low-level forensic noise artifacts extracted via Steganalysis Rich Model (SRM) filters using a ResNet like 4-stage backbone, to accurately classify the input image into one of four common manipulation types: Copy-Move, Splicing, Removal (Inpainting), or Enhancement. In the second stage, the classified image is directed to a specialized Conditional Diffusion Model (CDM), which treats the localization task as an image-to-mask generation problem. These CDMs are conditioned on the input image, and are trained with a modified loss function incorporating both Mean Squared Error (MSE) and Intersection over Union (IoU) to better handle smaller manipulation masks and encourage better spatial structure. Evaluated on the DF2023 dataset, the manipulation classifier achieves an accuracy of 89% and localization CDMs achieve average IoU of 0.70 and F1-Score of 0.77. The system demonstrates competitive performance on benchmark datasets, confirming the efficacy of a modular, generative approach for tackling the complexity of image forgeries.
Keywords:
Image Manipulation Localization
Conditional Diffusion
Dual-Stream Classifier
Image Forensics
Journal
IF:
3
Papers:
2.0W
Citations:
3.2W
Organization
No organization information available
Cited Papers
DH-GAN: Image manipulation localization via a dual homology-aware generative adversarial network.png)
PATTERN RECOGNITION
PATTERN RECOGNITIONIF7.6
PREAI
ShareSaveA Weakly Supervised Multi-Stream Feature Enhancement Network for Image Manipulation Detection and LocalizationIEEE Access
OAAI
IEEE AccessIF3.6
OAAIShareSaveShareSaveUnravelling Digital Forgeries: A Systematic Survey on Image Manipulation Detection and Localization
PREAI
ShareSaveShareSaveLearning Traces by Yourself: Blind Image Forgery Localization via Anomaly Detection With ViT-VAE
PREAI
ShareSave
ShareSave
View more