scholarly journals Hybrid reference-based Video Source Identification

Sensors ◽  
2019 ◽  
Vol 19 (3) ◽  
pp. 649 ◽  
Author(s):  
Massimo Iuliani ◽  
Marco Fontani ◽  
Dasara Shullani ◽  
Alessandro Piva
Keyword(s):  
Social Media ◽  
Source Identification ◽  
State Of The Art ◽  
Visual Media ◽  
Multimedia Forensics ◽  
Reference Pattern ◽  
Still Images ◽  
Video Source ◽  
Video Frames ◽  
Social Media Platforms

Millions of users share images and videos generated by mobile devices with different profiles on social media platforms. When publishing illegal content, they prefer to use anonymous profiles. Multimedia Forensics allows us to determine whether videos or images have been captured with the same device, and thus, possibly, by the same person. Currently, the most promising technology to achieve this task exploits unique traces left by the camera sensor into the visual content. However, image and video source identification are still treated separately from one another. This approach is limited and anachronistic, if we consider that most of the visual media are today acquired using smartphones that capture both images and videos. In this paper we overcome this limitation by exploring a new approach that synergistically exploits images and videos to study the device from which they both come. Indeed, we prove it is possible to identify the source of a digital video by exploiting a reference sensor pattern noise generated from still images taken by the same device. The proposed method provides performance comparable with or even better than the state-of-the-art, where a reference pattern is estimated from video frames. Finally, we show that this strategy is effective even in the case of in-camera digitally stabilized videos, where a non-stabilized reference is not available, thus solving the limitations of the current state-of-the-art. We also show how this approach allows us to link social media profiles containing images and videos captured by the same sensor.

Redefining #YourAverageMuslim woman: Muslim female digital activism on social media

2019 ◽  
Vol 12 (2) ◽  
pp. 213-233 ◽  
Author(s):  
Inaash Islam
Keyword(s):  
Social Media ◽  
Visual Representations ◽  
Muslim Women ◽  
Visual Media ◽  
Muslim Woman ◽  
Complex Relationship ◽  
Digital Activism ◽  
Women In Art ◽  
Controlling Images ◽  
Social Media Platforms

Orientalist discourses have largely shaped how Muslim women have come to be represented in western visual media as oppressed, subjugated or foreign. However, with the advent of social media platforms, Muslim women are utilizing social media spaces to rearticulate the controlling images promulgated through orientalist narratives. This article examines the complex relationship visual media shares with Muslim women and demonstrates that the lens of orientalism continues to structure the imaginaries that shape visual representations of Muslim women in art, news and film. This article addresses how visual platforms and social media spaces such as YouTube are being utilized by Muslim women to undertake digital activism that seeks to subvert essentialist narratives. At the centre of this discussion is YouTuber Dina Tokio’s (2017) documentary, titled ‘#YourAverageMuslim’, which tackles western preconceived notions, and instead offers a redefined version of the ‘Muslim woman’ predicated on resisting three narratives: (1) Muslim-Woman-As-Oppressed, (2) Muslim-Woman-As-Subjugated and (3) Muslim-Woman-As-Foreign-Other. This documentary clearly demonstrates how Muslim women are using social media platforms in specific ways to shape the discourses around Muslim women. In doing so they are demonstrating their agentic capabilities, taking control of their representations, and speaking for themselves instead of being spoken for by others.

scholarly journals Detection of Manipulated Face Videos over Social Networks: A Large-Scale Study

2021 ◽  
Vol 7 (10) ◽  
pp. 193
Author(s):  
Federico Marcon ◽  
Cecilia Pasquini ◽  
Giulia Boato
Keyword(s):  
Large Scale ◽  
State Of The Art ◽  
Forensic Analysis ◽  
General Purpose ◽  
Fine Tuning ◽  
Specific Technique ◽  
Multimedia Forensics ◽  
Shared Data ◽  
Social Media Platforms ◽  
The Web

The detection of manipulated videos represents a highly relevant problem in multimedia forensics, which has been widely investigated in the last years. However, a common trait of published studies is the fact that the forensic analysis is typically applied on data prior to their potential dissemination over the web. This work addresses the challenging scenario where manipulated videos are first shared through social media platforms and then are subject to the forensic analysis. In this context, a large scale performance evaluation has been carried out involving general purpose deep networks and state-of-the-art manipulated data, and studying different effects. Results confirm that a performance drop is observed in every case when unseen shared data are tested by networks trained on non-shared data; however, fine-tuning operations can mitigate this problem. Also, we show that the output of differently trained networks can carry useful forensic information for the identification of the specific technique used for visual manipulation, both for shared and non-shared data.

scholarly journals Towards Combating Pandemic-Related Misinformation in Social Media

2021 ◽  
pp. 140-158
Author(s):  
Isa Inuwa-Dutse
Keyword(s):  
Social Media ◽  
Public Discourse ◽  
Preventive Measures ◽  
State Of The Art ◽  
Fake News ◽  
Social Distancing ◽  
The Public ◽  
Toxic Impact ◽  
Online Social Media ◽  
Social Media Platforms

Conventional preventive measures during pandemics include social distancing and lockdown. Such measures in the time of social media brought about a new set of challenges – vulnerability to the toxic impact of online misinformation is high. A case in point is COVID-19. As the virus propagates, so does the associated misinformation and fake news about it leading to an infodemic. Since the outbreak, there has been a surge of studies investigating various aspects of the pandemic. Of interest to this chapter are studies centering on datasets from online social media platforms where the bulk of the public discourse happens. The main goal is to support the fight against negative infodemic by (1) contributing a diverse set of curated relevant datasets; (2) offering relevant areas to study using the datasets; and (3) demonstrating how relevant datasets, strategies, and state-of-the-art IT tools can be leveraged in managing the pandemic.

scholarly journals Self-Supervised Video Representation Learning with Space-Time Cubic Puzzles

2019 ◽  
Vol 33 ◽  
pp. 8545-8552 ◽  
Author(s):  
Dahun Kim ◽  
Donghyeon Cho ◽  
In So Kweon
Keyword(s):  
Action Recognition ◽  
Large Scale ◽  
State Of The Art ◽  
Representation Learning ◽  
Space Time ◽  
Temporal Relation ◽  
Still Images ◽  
Video Frames ◽  
Spatio Temporal ◽  
The Cost

Self-supervised tasks such as colorization, inpainting and zigsaw puzzle have been utilized for visual representation learning for still images, when the number of labeled images is limited or absent at all. Recently, this worthwhile stream of study extends to video domain where the cost of human labeling is even more expensive. However, the most of existing methods are still based on 2D CNN architectures that can not directly capture spatio-temporal information for video applications. In this paper, we introduce a new self-supervised task called as Space-Time Cubic Puzzles to train 3D CNNs using large scale video dataset. This task requires a network to arrange permuted 3D spatio-temporal crops. By completing Space-Time Cubic Puzzles, the network learns both spatial appearance and temporal relation of video frames, which is our final goal. In experiments, we demonstrate that our learned 3D representation is well transferred to action recognition tasks, and outperforms state-of-the-art 2D CNN-based competitors on UCF101 and HMDB51 datasets.

scholarly journals Media forensics on social media platforms: a survey

2021 ◽  
Vol 2021 (1) ◽  
Author(s):  
Cecilia Pasquini ◽  
Irene Amerini ◽  
Giulia Boato
Keyword(s):  
Social Media ◽  
Digital Media ◽  
Visual Information ◽  
Open Systems ◽  
Forensic Analysis ◽  
Provenance Analysis ◽  
Multimedia Forensics ◽  
Media Forensics ◽  
Media Objects ◽  
Social Media Platforms

AbstractThe dependability of visual information on the web and the authenticity of digital media appearing virally in social media platforms has been raising unprecedented concerns. As a result, in the last years the multimedia forensics research community pursued the ambition to scale the forensic analysis to real-world web-based open systems. This survey aims at describing the work done so far on the analysis of shared data, covering three main aspects: forensics techniques performing source identification and integrity verification on media uploaded on social networks, platform provenance analysis allowing to identify sharing platforms, and multimedia verification algorithms assessing the credibility of media objects in relation to its associated textual information. The achieved results are highlighted together with current open issues and research challenges to be addressed in order to advance the field in the next future.

Seeing Syria

2017 ◽  
Vol 10 (2-3) ◽  
pp. 133-158 ◽  
Author(s):  
Rhys Crilley
Keyword(s):  
Social Media ◽  
English Language ◽  
Visual Media ◽  
Islamic State ◽  
The Press ◽  
Social Media Platforms ◽  
National Coalition ◽  
Use Of Social Media ◽  
Syrian Revolution ◽  
Syrian Conflict

The use of social media by groups such as the self-proclaimed Islamic State has been the focus of the press, politicians and scholars, but relatively little attention has been paid to how other actors involved in the Syrian conflict have been using social media platforms. In this article, I address this gap by analyzing how the National Coalition of Syrian Revolution and Opposition Forces has used Facebook. Specifically, I focus on the main narrative themes emerging in 1,174 posts uploaded to the Coalition’s English language Facebook page between November 2012 and March 2015. Recognizing that visual media play an important factor in communicating narratives of conflict, the paper also analyzes 280 sets of images of war posted during this time. I argue that these images construct a visuality focused on ‘the pain of others’ (Sontag 2004) that makes those affected, uprooted, injured, and killed by the conflict in Syria highly visible. In making this claim, I explore how this visuality of suffering has evolved over the course of the conflict.

The effectiveness of audio-visual media and social media platforms as components of blended learning

2020 ◽  
Vol 9 (4) ◽  
pp. 177-195
Author(s):  
Michael R. Brett ◽  
Keyword(s):  
South Africa ◽  
Social Media ◽  
Blended Learning ◽  
Visual Media ◽  
University Education ◽  
Education Students ◽  
Tertiary Institutions ◽  
Course Content ◽  
Social Media Platforms ◽  
Traditional Assessments

South Africa currently has 90 million cellphone connections and 4G bandwidth is accessible to 75% of the population. Audio-visual media, such as videos, can be used to enhance teaching as the use of multimedia is a key component of blended learning. In total, 92 fourth-year university education students were surveyed to determine their response to video-based assessments. Of the students surveyed, 92% believed that videos assisted their understanding of the course content, 78.5% believed that video-based assessments were less difficult than traditional assessments and 89% intend using audio-visual media in their own classrooms. Significantly, 88.7% believe that such media should be used at least once a week. In addition, once they graduate, 63% of participants intend using social media platforms to communicate with learners. The study suggests that a greater emphasis needs to be placed on blended learning, both in schools and tertiary institutions in South Africa.

scholarly journals Visual media analysis for Instagram and other online platforms

2021 ◽  
Vol 8 (1) ◽  
pp. 205395172110223
Author(s):  
Richard Rogers
Keyword(s):  
Social Media ◽  
Visual Media ◽  
Media Analysis ◽  
Image Search ◽  
Social Media Platform ◽  
Online Platforms ◽  
The Social ◽  
Social Media Platforms ◽  
Media Platform ◽  
Initial Focus

Instagram is currently the social media platform most associated with online images (and their analysis), but images from other platforms also can be collected and grouped, arrayed by similarity, stacked, matched, stained, labelled, depicted as network, placed side by side and otherwise analytically displayed. In the following, the initial focus is on Instagram, together with certain schools of thought such as Instagramism and Instagrammatics for its aesthetic and visual cultural study. Building on those two approaches, it subsequently focuses on other web and social media platforms, such as Google Image Search, Twitter, Facebook and 4chan. It provides demonstrations of how querying techniques create online image collections, and how these sets are analytically grouped through arrangements collectively referred to as metapictures.

scholarly journals Detection of Cyberbullying Through BERT and Weighted Ensemble of Classifiers

2022 ◽  
Author(s):  
Christopher Graney-Ward ◽  
Biju Issac ◽  
LIDA KETSBAIA ◽  
Seibu Mary Jacob
Keyword(s):  
Machine Learning ◽  
Social Media ◽  
State Of The Art ◽  
Particle Swarm Optimisation ◽  
Ensemble Model ◽  
Ensemble Of Classifiers ◽  
Weight Decay ◽  
Data Representations ◽  
Online Behaviour ◽  
Social Media Platforms

Due to the recent popularity and growth of social media platforms such as Facebook and Twitter, cyberbullying is becoming more and more prevalent. The current research on cyberbullying and the NLP techniques being used to classify this kind of online behaviour was initially studied. This paper discusses the experimentation with combined Twitter datasets by Maryland and Cornell universities using different classification approaches like classical machine learning, RNN, CNN, and pretrained transformer-based classifiers. A state of the art (SOTA) solution was achieved by optimising BERTweet on a Onecycle policy with a Decoupled weight decay optimiser (AdamW), improving the previous F1-score by up to 8.4%, resulting in 64.8% macro F1. Particle Swarm Optimisation was later used to optimise the ensemble model. The ensemble developed from the optimised BERTweet model and a collection of models with varying data representations, outperformed the standalone BERTweet model by 0.53% resulting in 65.33% macro F1 for TweetEval dataset and by 0.55% for combined datasets, resulting in 68.1% macro F1.

scholarly journals Detection of Cyberbullying Through BERT and Weighted Ensemble of Classifiers

2022 ◽  
Author(s):  
Christopher Graney-Ward ◽  
Biju Issac ◽  
LIDA KETSBAIA ◽  
Seibu Mary Jacob
Keyword(s):  
Machine Learning ◽  
Social Media ◽  
State Of The Art ◽  
Particle Swarm Optimisation ◽  
Ensemble Model ◽  
Ensemble Of Classifiers ◽  
Weight Decay ◽  
Data Representations ◽  
Online Behaviour ◽  
Social Media Platforms

Due to the recent popularity and growth of social media platforms such as Facebook and Twitter, cyberbullying is becoming more and more prevalent. The current research on cyberbullying and the NLP techniques being used to classify this kind of online behaviour was initially studied. This paper discusses the experimentation with combined Twitter datasets by Maryland and Cornell universities using different classification approaches like classical machine learning, RNN, CNN, and pretrained transformer-based classifiers. A state of the art (SOTA) solution was achieved by optimising BERTweet on a Onecycle policy with a Decoupled weight decay optimiser (AdamW), improving the previous F1-score by up to 8.4%, resulting in 64.8% macro F1. Particle Swarm Optimisation was later used to optimise the ensemble model. The ensemble developed from the optimised BERTweet model and a collection of models with varying data representations, outperformed the standalone BERTweet model by 0.53% resulting in 65.33% macro F1 for TweetEval dataset and by 0.55% for combined datasets, resulting in 68.1% macro F1.

Sign in / Sign up

Export Citation Format

Share Document