Deep Multimodal Image-Repurposing Detection

Nefarious actors on social media and other platforms often spread rumors and falsehoods through images whose metadata (e.g., captions) have been modified to provide visual substantiation of the rumor/falsehood. This type of modification is referred to as image repurposing, in which often an unmanipulated image is published along with incorrect or manipulated metadata to serve the actor's ulterior motives. We present the Multimodal Entity Image Repurposing (MEIR) dataset, a substantially challenging dataset over that which has been previously available to support research into image repurposing detection. The new dataset includes location, person, and organization manipulations on real-world data sourced from Flickr. We also present a novel, end-to-end, deep multimodal learning model for assessing the integrity of an image by combining information extracted from the image with related information from a knowledge base. The proposed method is compared against state-of-the-art techniques on existing datasets as well as MEIR, where it outperforms existing methods across the board, with AUC improvement up to 0.23.

[1]  Jeffrey Dean,et al.  Distributed Representations of Words and Phrases and their Compositionality , 2013, NIPS.

[2]  Sungyong Seo,et al.  CSI: A Hybrid Deep Model for Fake News Detection , 2017, CIKM.

[3]  Zulfiqar Habib,et al.  Copy-move and splicing image forgery detection and localization techniques: a review , 2017 .

[4]  Yongdong Zhang,et al.  MCG-ICT at MediaEval 2015: Verifying Multimedia Use with a Two-Level Classification Model , 2015, MediaEval.

[5]  Chong-sun Kim Canonical Analysis of Several Sets of Variables , 1973 .

[6]  Mandeep Kaur,et al.  Copy Move Tampering Detection Techniques : A Review , 2016 .

[7]  Wael Abd-Almageed,et al.  Multimedia Semantic Integrity Assessment Using Joint Embedding Of Images And Text , 2017, ACM Multimedia.

[8]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[9]  Yongdong Zhang,et al.  Novel Visual and Statistical Image Features for Microblogs News Verification , 2017, IEEE Transactions on Multimedia.

[10]  Sanjoy Dasgupta,et al.  Experiments with Random Projection , 2000, UAI.

[11]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[12]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[13]  Yiannis Kompatsiaris,et al.  Verifying Multimedia Use at MediaEval 2016 , 2015, MediaEval.

[14]  Heikki Mannila,et al.  Random projection in dimensionality reduction: applications to image and text data , 2001, KDD '01.

[15]  Yiannis Kompatsiaris,et al.  Web and Social Media Image Forensics for News Professionals , 2021, SMN@ICWSM.

[16]  Yongdong Zhang,et al.  Multimodal Fusion with Recurrent Neural Networks for Rumor Detection on Microblogs , 2017, ACM Multimedia.

[17]  Leo Breiman,et al.  Classification and Regression Trees , 1984 .

[18]  Mohamed Deriche,et al.  A bibliography of pixel-based blind image forgery detection techniques , 2015, Signal Process. Image Commun..

[19]  Carey E. Priebe,et al.  Generalized Canonical Correlation Analysis for Disparate Data Fusion , 2013, Pattern Recognit. Lett..

[20]  Christopher D. Manning,et al.  Incorporating Non-local Information into Information Extraction Systems by Gibbs Sampling , 2005, ACL.

[21]  Regina M. Marchi With Facebook, Blogs, and Fake News, Teens Reject Journalistic “Objectivity” , 2012 .

[22]  Marina Del Rey,et al.  Deep Matching and Validation Network: An End-to-End Solution to Constrained Image Splicing Localization and Detection , 2017, ACM Multimedia.

[23]  Michael S. Bernstein,et al.  ImageNet Large Scale Visual Recognition Challenge , 2014, International Journal of Computer Vision.

[24]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[25]  Carey E. Priebe,et al.  Generalized canonical correlation analysis for classification , 2013, J. Multivar. Anal..