Localization of JPEG Double Compression Through Multi-domain Convolutional Neural Networks

Amerini, Irene; Uricchio, Tiberio; Ballan, Lamberto; Caldelli, Roberto

doi:10.1109/cvprw.2017.233

Cited by 116 publications

(60 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Early work detected resampling artifacts [28,20] by finding periodic correlations between nearby pixels. There has also been work that detects inconsistent quantization [4], double-JPEG artifacts [8,5], and geometric inconsistencies [26]. However, the operations performed by interactive image editing tools are often complex, and can be difficult to model.…”

Section: Hand-defined Manipulation Cuesmentioning

confidence: 99%

Detecting Photoshopped Faces by Scripting Photoshop

Wang

Zhang

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

109

View full text Add to dashboard Cite

a) Manipulated photo (b) Detected manipulations (c) Suggested "undo" (d) Original photo Figure 1: Given an input face (a), our tool can detect that the face has been warped with the Face-Aware Liquify tool from Photoshop, predict where the face has been warped (b), and attempt to "undo" the warp (c) and recover the original image (d). AbstractMost malicious photo manipulations are created using standard image editing tools, such as Adobe R Photoshop R . We present a method for detecting one very popular Photoshop manipulation -image warping applied to human faces -using a model trained entirely using fake images that were automatically generated by scripting Photoshop itself. We show that our model outperforms humans at the task of recognizing manipulated images, can predict the specific location of edits, and in some cases can be used to "undo" a manipulation to reconstruct the original, unedited image. We demonstrate that the system can be successfully applied to real, artist-created image manipulations.

show abstract

Section: Hand-defined Manipulation Cuesmentioning

confidence: 99%

Detecting Photoshopped Faces by Scripting Photoshop

Wang

Zhang

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

109

View full text Add to dashboard Cite

show abstract

“…Combining forensic methodologies and recent advancements established by deep learning techniques in computer vision, some researchers [29,7,2,1] have proposed to learn camera identification features by using convolutional neural networks (CNN). The advantage of CNN is that they are capable of learning classification features directly from data, hence, they adaptively learn the cumulative traces induced by camera components.…”

Section: Introductionmentioning

confidence: 99%

MultiMedia Modeling

2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

The task of verifying the originality and authenticity of images puts numerous constraints on tampering detection algorithms. Since most images are acquired on the internet, there is a significant probability that they have undergone transformations such as compression, noising, resizing and/or filtering, both before and after the possible alteration. Therefore, it is essential to improve the robustness of tampered image detection algorithms for such manipulations. As compression is the most common type of post-processing, we propose in our work a robust framework against this particular transformation. Our experiments on benchmark datasets show the contribution of our proposal for camera model identification and image tampering detection compared to recent literature approaches.

show abstract

“…CNN and machine learning are used extensively in many areas such as image classification and object detection and recently also in image forensics. CNN have also been employed to detect image manipulations by revealing single or double JPEG compressions [12], [13], [14] and to perform source camera identification [15], [16]. Deep learning has also been used very recently in [17] for classifying four types of global processing applied to an image namely low-pass filtering (blurring), high-pass filtering (sharpening), denoising (content adaptive low-pass filtering), and tonal adjustment.…”

Section: Introductionmentioning

confidence: 99%

PRNU-based Image Classification of Origin Social Network with CNN

Caldelli

Amerini

2018

2018 26th European Signal Processing Conference (EUSIPCO)

Self Cite

View full text Add to dashboard Cite

A huge amount of images are continuously shared on social networks (SNs) daily and, in most of cases, it is very difficult to reliably establish the SN of provenance of an image when it is recovered from a hard disk, a SD card or a smartphone memory. During an investigation, it could be crucial to be able to distinguish images coming directly from a photo-camera with respect to those downloaded from a social network and possibly, in this last circumstance, determining which is the SN among a defined group. It is well known that each SN leaves peculiar traces on each content during the upload-download process; such traces can be exploited to make image classification. In this work, the idea is to use the PRNU, embedded in every acquired images, as the "carrier" of the particular SN traces which diversely modulate the PRNU. We demonstrate, in this paper, that SNmodulated noise residual can be adopted as a feature to detect the social network of origin by means of a trained convolutional neural network (CNN).

show abstract

Localization of JPEG Double Compression Through Multi-domain Convolutional Neural Networks

Cited by 116 publications

References 18 publications

Detecting Photoshopped Faces by Scripting Photoshop

Detecting Photoshopped Faces by Scripting Photoshop

MultiMedia Modeling

PRNU-based Image Classification of Origin Social Network with CNN

Contact Info

Product

Resources

About