CASIA OpenIR  > 模式识别国家重点实验室  > 多媒体计算与图形学
Distinguishing Between Natural and Computer-Generated Images Using Convolutional Neural Networks
Weize Quan; Kai Wang; Dong-Ming Yan; Xiaopeng Zhang
Source PublicationIEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
2018-05-07
Volume13Issue:11Pages:2772 - 2787
AbstractDistinguishing between natural images (NIs) and computer-generated (CG) images by naked human eyes is difficult. In this paper, we propose an effective method based on a convolutional neural network (CNN) for this fundamental image forensic problem. Having observed the rather limited performance of training existing CCNs from scratch or fine-tuning pre-trained network, we design and implement a new and appropriate network with two cascaded convolutional layers at the bottom of a CNN. Our network can be easily adjusted to accommodate different sizes of input image patches while maintaining a fixed depth, a stable structure of CNN, and a good forensic performance. Considering the complexity of training CNNs and the specific requirement of image forensics, we introduce the so-called local-to-global strategy in our proposed network. Our CNN derives a forensic decision on local patches, and a global decision on a full-sized image can be easily obtained via simple majority voting. This strategy can also be used to improve the performance of existing methods that are based on hand-crafted features. Experimental results show that our method outperforms existing methods, especially in a challenging forensic scenario with NIs and CG images of heterogeneous origins. Our method also has good robustness against typical post-processing operations, such as resizing and JPEG compression. Unlike previous attempts to use CNNs for image forensics, we try to understand what our CNN has learned about the differences between NIs and CG images with the aid of adequate and advanced visualization tools.; Distinguishing between natural images (NIs) and computer-generated (CG) images by naked human eyes is difficult. In this paper, we propose an effective method based on a convolutional neural network (CNN) for this fundamental image forensic problem. Having observed the rather limited performance of training existing CCNs from scratch or fine-tuning pre-trained network, we design and implement a new and appropriate network with two cascaded convolutional layers at the bottom of a CNN. Our network can be easily adjusted to accommodate different sizes of input image patches while maintaining a fixed depth, a stable structure of CNN, and a good forensic performance. Considering the complexity of training CNNs and the specific requirement of image forensics, we introduce the so-called local-to-global strategy in our proposed network. Our CNN derives a forensic decision on local patches, and a global decision on a full-sized image can be easily obtained via simple majority voting. This strategy can also be used to improve the performance of existing methods that are based on hand-crafted features. Experimental results show that our method outperforms existing methods, especially in a challenging forensic scenario with NIs and CG images of heterogeneous origins. Our method also has good robustness against typical post-processing operations, such as resizing and JPEG compression. Unlike previous attempts to use CNNs for image forensics, we try to understand what our CNN has learned about the differences between NIs and CG images with the aid of adequate and advanced visualization tools.
KeywordImage Forensics Natural Image Computergenerated Image Convolutional Neural Network Robustness Localto- Global Strategy Visualization
DOI10.1109/TIFS.2018.2834147
Citation statistics
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/21683
Collection模式识别国家重点实验室_多媒体计算与图形学
Corresponding AuthorDong-Ming Yan
Affiliation1.National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences
2.the University of Chinese Academy of Sciences
3.University Grenoble Alpes, CNRS, Grenoble INP, GIPSA-lab
Recommended Citation
GB/T 7714
Weize Quan,Kai Wang,Dong-Ming Yan,et al. Distinguishing Between Natural and Computer-Generated Images Using Convolutional Neural Networks[J]. IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY,2018,13(11):2772 - 2787.
APA Weize Quan,Kai Wang,Dong-Ming Yan,&Xiaopeng Zhang.(2018).Distinguishing Between Natural and Computer-Generated Images Using Convolutional Neural Networks.IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY,13(11),2772 - 2787.
MLA Weize Quan,et al."Distinguishing Between Natural and Computer-Generated Images Using Convolutional Neural Networks".IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY 13.11(2018):2772 - 2787.
Files in This Item: Download All
File Name/Size DocType Version Access License
2018_TIFS_iunacg.pdf(8950KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Weize Quan]'s Articles
[Kai Wang]'s Articles
[Dong-Ming Yan]'s Articles
Baidu academic
Similar articles in Baidu academic
[Weize Quan]'s Articles
[Kai Wang]'s Articles
[Dong-Ming Yan]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Weize Quan]'s Articles
[Kai Wang]'s Articles
[Dong-Ming Yan]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 2018_TIFS_iunacg.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.