End4: End-to-end Denoising Diffusion for Diffusion-Based Inpainting Detection

2025-09-16

AI Summary

To address the malicious misuse of diffusion-based image inpainting and the poor generalization of existing detection methods, this paper proposes an end-to-end denoising diffusion detection framework. The framework introduces a denoising reconstruction module with latent-space alignment to ensure consistency between detection and generative processes; it further designs a Scale-Aware Pyramid Fusion Module (SPFM) that integrates attention-guided multi-scale feature extraction to significantly enhance local discriminability within inpainted regions. The method demonstrates strong robustness across diverse masking patterns and perturbations. Moreover, we construct the first benchmark dataset covering five representative masked region categories. Experiments show that our approach achieves substantial improvements over state-of-the-art methods in detection accuracy, cross-domain generalization, and resilience to adversarial interference.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Adversarial Learning & Robustness
• Natural Language Processing: Safety and Robustness

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Security and Privacy: Large-scale security measurements

Abstract

The powerful generative capabilities of diffusion models have significantly advanced the field of image synthesis, enhancing both full image generation and inpainting-based image editing. Despite their remarkable advancements, diffusion models also raise concerns about potential misuse for malicious purposes. However, existing approaches struggle to identify images generated by diffusion-based inpainting models, even when similar inpainted images are included in their training data. To address this challenge, we propose a novel detection method based on End-to-end denoising diffusion (End4). Specifically, End4 designs a denoising reconstruction model to improve the alignment degree between the latent spaces of the reconstruction and detection processes, thus reconstructing features that are more conducive to detection. Meanwhile, it leverages a Scale-aware Pyramid-like Fusion Module (SPFM) that refines local image features under the guidance of attention pyramid layers at different scales, enhancing feature discriminability. Additionally, to evaluate detection performance on inpainted images, we establish a comprehensive benchmark comprising images generated from five distinct masked regions. Extensive experiments demonstrate that our End4 effectively generalizes to unseen masking patterns and remains robust under various perturbations. Our code and dataset will be released soon.

Problem

• Detecting images generated by diffusion-based inpainting models
• Improving feature alignment between reconstruction and detection processes
• Establishing benchmark for inpainting detection performance evaluation

Innovation

• End-to-end denoising diffusion reconstruction model
• Scale-aware Pyramid-like Fusion Module
• Comprehensive benchmark with five masked regions