IRL-Net: Inpainted Region Localization Network via Spatial Attention
Identifying manipulated regions in images is a challenging task due to the existence of very accurate image inpainting techniques leaving almost unnoticeable traces in tampered regions. These image inpainting methods can be used for multiple purposes (e.g., removing objects, reconstructing corrupted...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10285582/ |
| _version_ | 1797649813860253696 |
|---|---|
| author | Amir Etefaghi Daryani Mahdieh Mirmahdi Ahmad Hassanpour Hatef Otroshi Shahreza Bian Yang Julian Fierrez |
| author_facet | Amir Etefaghi Daryani Mahdieh Mirmahdi Ahmad Hassanpour Hatef Otroshi Shahreza Bian Yang Julian Fierrez |
| author_sort | Amir Etefaghi Daryani |
| collection | DOAJ |
| description | Identifying manipulated regions in images is a challenging task due to the existence of very accurate image inpainting techniques leaving almost unnoticeable traces in tampered regions. These image inpainting methods can be used for multiple purposes (e.g., removing objects, reconstructing corrupted areas, eliminating various types of distortion, etc.) makes creating forensic detectors for image manipulation an extremely difficult and time-consuming procedure. The aim of this paper is to localize the tampered regions manipulated by image inpainting methods. To do this, we propose a novel CNN-based deep learning model called IRL-Net which includes three main modules: Enhancement, Encoder, and Decoder modules. To evaluate our method, three image inpainting methods have been used to reconstruct the missed regions in two face and scene image datasets. We perform both qualitative and quantitative evaluations on the generated datasets. Experimental results demonstrate that our method outperforms previous learning-based manipulated region detection methods and generates realistic and semantically plausible images. We also provide the implementation of the proposed approach to support reproducible research via <uri>https://github.com/amiretefaghi/IRL-Net</uri>. |
| first_indexed | 2024-03-11T15:51:26Z |
| format | Article |
| id | doaj.art-ebeaeffa69fa44d7b029678fa7059663 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-11T15:51:26Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-ebeaeffa69fa44d7b029678fa70596632023-10-25T23:01:02ZengIEEEIEEE Access2169-35362023-01-011111567711568710.1109/ACCESS.2023.332454110285582IRL-Net: Inpainted Region Localization Network via Spatial AttentionAmir Etefaghi Daryani0https://orcid.org/0000-0002-8692-7047Mahdieh Mirmahdi1Ahmad Hassanpour2https://orcid.org/0000-0002-3936-2223Hatef Otroshi Shahreza3https://orcid.org/0000-0002-8199-0098Bian Yang4Julian Fierrez5https://orcid.org/0000-0002-6343-5656Department of Electrical Engineering, Amirkabir University of Technology, Tehran, IranFaculty of Computer Engineering, University of Isfahan, Isfahan, IranDepartment of Information Security and Communication Technology, Norwegian University of Science and Technology (NTNU), Gjøvik, NorwayBiometrics Security and Privacy Group, Idiap Research Institute, Martigny, SwitzerlandDepartment of Information Security and Communication Technology, Norwegian University of Science and Technology (NTNU), Gjøvik, NorwaySchool of Engineering, Universidad Autónoma de Madrid, Madrid, SpainIdentifying manipulated regions in images is a challenging task due to the existence of very accurate image inpainting techniques leaving almost unnoticeable traces in tampered regions. These image inpainting methods can be used for multiple purposes (e.g., removing objects, reconstructing corrupted areas, eliminating various types of distortion, etc.) makes creating forensic detectors for image manipulation an extremely difficult and time-consuming procedure. The aim of this paper is to localize the tampered regions manipulated by image inpainting methods. To do this, we propose a novel CNN-based deep learning model called IRL-Net which includes three main modules: Enhancement, Encoder, and Decoder modules. To evaluate our method, three image inpainting methods have been used to reconstruct the missed regions in two face and scene image datasets. We perform both qualitative and quantitative evaluations on the generated datasets. Experimental results demonstrate that our method outperforms previous learning-based manipulated region detection methods and generates realistic and semantically plausible images. We also provide the implementation of the proposed approach to support reproducible research via <uri>https://github.com/amiretefaghi/IRL-Net</uri>.https://ieeexplore.ieee.org/document/10285582/Image forensicsimage inpaintingimage manipulation detection |
| spellingShingle | Amir Etefaghi Daryani Mahdieh Mirmahdi Ahmad Hassanpour Hatef Otroshi Shahreza Bian Yang Julian Fierrez IRL-Net: Inpainted Region Localization Network via Spatial Attention IEEE Access Image forensics image inpainting image manipulation detection |
| title | IRL-Net: Inpainted Region Localization Network via Spatial Attention |
| title_full | IRL-Net: Inpainted Region Localization Network via Spatial Attention |
| title_fullStr | IRL-Net: Inpainted Region Localization Network via Spatial Attention |
| title_full_unstemmed | IRL-Net: Inpainted Region Localization Network via Spatial Attention |
| title_short | IRL-Net: Inpainted Region Localization Network via Spatial Attention |
| title_sort | irl net inpainted region localization network via spatial attention |
| topic | Image forensics image inpainting image manipulation detection |
| url | https://ieeexplore.ieee.org/document/10285582/ |
| work_keys_str_mv | AT amiretefaghidaryani irlnetinpaintedregionlocalizationnetworkviaspatialattention AT mahdiehmirmahdi irlnetinpaintedregionlocalizationnetworkviaspatialattention AT ahmadhassanpour irlnetinpaintedregionlocalizationnetworkviaspatialattention AT hatefotroshishahreza irlnetinpaintedregionlocalizationnetworkviaspatialattention AT bianyang irlnetinpaintedregionlocalizationnetworkviaspatialattention AT julianfierrez irlnetinpaintedregionlocalizationnetworkviaspatialattention |