Progressive Reverse Attention Network for image inpainting detection and localization
Image inpainting is originally presented to restore damaged image areas, but it might be maliciously used for object removal that change image semantic content. This easily leads to serious public confidence crises. Up to present, image inpainting forensics works have achieved remarkable results, bu...
Saved in:
| Published in: | Computer vision and image understanding Vol. 259; p. 104407 |
|---|---|
| Main Authors: | , , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Elsevier Inc
01.09.2025
|
| Subjects: | |
| ISSN: | 1077-3142 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Image inpainting is originally presented to restore damaged image areas, but it might be maliciously used for object removal that change image semantic content. This easily leads to serious public confidence crises. Up to present, image inpainting forensics works have achieved remarkable results, but they usually ignore or fail to capture subtle artifacts near object boundary, resulting in inaccurate object mask localization. To address this issue, we propose a Progressive Reverse Attention Network (PRA-Net) for image inpainting detection and localization. Different from the traditional convolutional neural networks (CNN) structure, PRA-Net follows an encoder and decoder architecture. The encoder leverages features at different scales with dense cross-connections to locate inpainted regions and generates global map with our designed multi-scale extraction module. A reverse attention module is used as the backbone of the decoder to progressively refine the details of predictions. Experimental results show that PRA-Net achieves accurate image inpainting localization and desirable robustness.
•We build an open dataset for image inpainting forensics by using four deep inpainting works and one traditional work. It can serve as the benchmark dataset for comparisons.•We propose PRA-Net, an encoder–decoder architecture for image inpainting localization, leveraging HR-Net to hierarchically extract multi-scale features and enable precise pixel-level localization.•We introduce a reverse attention module and propose an edge supervision function based on dice loss, which enforces the model to focus on inpainted region boundaries for better performance.•We design a MSEM that enhances the model’s receptive field without down-sampling. It considers feature fusion at various scales, and provides excellent prior estimation for progressive decoder. |
|---|---|
| ISSN: | 1077-3142 |
| DOI: | 10.1016/j.cviu.2025.104407 |