scholarly journals Fighting Deepfakes Using Body Language Analysis

2021 ◽  
Author(s):  
Robail Yasrab ◽  
Wanqi Jiang ◽  
Adnan Riaz
Keyword(s):  
Open Source Software ◽  
Body Language ◽  
Classification Model ◽  
Upper Body ◽  
Final Model ◽  
The Public ◽  
Ulterior Motive ◽  
Language Analysis ◽  
Barrier To Entry ◽  
Lstm Network

Recent improvements in deepfake creation made deepfake videos more realistic. Open-source software has also made deepfake creation more accessible, which reduces the barrier to entry for deepfake creation. This could pose a threat to the public privacy. It is a potential danger if the deepfake creation techniques are used by people with an ulterior motive to produce deepfake videos of world leaders to disrupt the order of the countries and the world. Research into automated detection for deepfaked media is therefore essential for public safety. We propose in this work the use of upper body language analysis for deepfake detection. Specifically, a many-to-one LSTM network was designed and trained as a classification model is trained for deepfake detection. Different models trained using various hyper-parameters to build a final model with benchmark accuracy. We achieve 94.39% accuracy on a test deepfake set. The experimental results show that upper body language can effectively provide identification and deepfake detection.

scholarly journals Fighting Deepfakes Using Body Language Analysis

Forecasting ◽  
2021 ◽  
Vol 3 (2) ◽  
pp. 303-321
Author(s):  
Robail Yasrab ◽  
Wanqi Jiang ◽  
Adnan Riaz
Keyword(s):  
Open Source Software ◽  
Detection Method ◽  
Body Language ◽  
Classification Model ◽  
Upper Body ◽  
Final Model ◽  
Ulterior Motive ◽  
Language Analysis ◽  
Barrier To Entry ◽  
Lstm Network

Recent improvements in deepfake creation have made deepfake videos more realistic. Moreover, open-source software has made deepfake creation more accessible, which reduces the barrier to entry for deepfake creation. This could pose a threat to the people’s privacy. There is a potential danger if the deepfake creation techniques are used by people with an ulterior motive to produce deepfake videos of world leaders to disrupt the order of countries and the world. Therefore, research into the automatic detection of deepfaked media is essential for public security. In this work, we propose a deepfake detection method using upper body language analysis. Specifically, a many-to-one LSTM network was designed and trained as a classification model for deepfake detection. Different models were trained by varying the hyperparameters to build a final model with benchmark accuracy. We achieved 94.39% accuracy on the deepfake test set. The experimental results showed that upper body language can effectively detect deepfakes.

scholarly journals Fighting Deepfakes Using Body Language Analysis

2021 ◽  
Author(s):  
Robail Yasrab ◽  
Wanqi Jiang ◽  
Adnan Riaz
Keyword(s):  
Open Source Software ◽  
Body Language ◽  
Classification Model ◽  
Upper Body ◽  
Final Model ◽  
Ulterior Motive ◽  
The People ◽  
Language Analysis ◽  
Barrier To Entry ◽  
Lstm Network

Recent improvements in deepfake creation have made deepfake videos more realistic. Moreover, open-source software has made deepfake creation more accessible, which reduces the barrier to entry for deepfake creation. This could pose a threat to the people’s privacy. There is a potential danger if the deepfake creation techniques are used by people with an ulterior motive to produce deepfake videos of world leaders to disrupt the order of countries and the world. Therefore, research into the automatic detection of deepfaked media is essential for public security. In this work, we propose a deepfake detection method using upper body language analysis. Specifically, a many-to-one LSTM network was designed and trained as a classification model for deepfake detection. Different models were trained by varying the hyperparameters to build a final model with benchmark accuracy. We achieved 94.39% accuracy on the deepfake test set. The experimental results showed that upper body language can effectively detect deepfakes.

scholarly journals Energy-Related Behaviour of Consumers from the Silesia Province (Poland)—Towards a Low-Carbon Economy

Energies ◽  
2021 ◽  
Vol 14 (8) ◽  
pp. 2218
Author(s):  
Sylwia Słupik ◽  
Joanna Kos-Łabędowicz ◽  
Joanna Trzęsiok
Keyword(s):  
Classification Model ◽  
Attitudes And Beliefs ◽  
Low Carbon ◽  
Low Carbon Economy ◽  
The Public ◽  
Second Stage ◽  
Initial Stage ◽  
Low Emission ◽  
Energy Behaviour ◽  
Carbon Economy

The issue of energy behaviour among Polish consumers, and especially the motives and attitudes they manifest, is relatively under-researched. This article attempts to identify individual attitudes and beliefs of energy consumers using the example of the residents of the province of Silesia (Poland). The authors conducted the expert segmentation of respondents in terms of their motivation for saving energy, based on the results of their proprietary survey. The second stage of the study involved using a classification model that allowed for the characterisation of the obtained groups. Psychological and financial factors were of greatest significance, which is confirmed by the results of other studies. Nonetheless, the obtained results explicitly indicate the specificity of the region, which requires transformation towards a low-emission economy. Despite the initial stage of changes both in the awareness of the consumers and the public interventions of the authorities, it should be emphasized that a majority of the respondents—at least to a basic extent—declared taking energy-saving measures. Financial motives are predominant among the respondents, although pro-environmental motives can also be noticed, which might translate into increased involvement and concern for the environment and climate.

KArSL

2021 ◽  
Vol 20 (1) ◽  
pp. 1-19
Author(s):  
Ala Addin I. Sidig ◽  
Hamzah Luqman ◽  
Sabri Mahmoud ◽  
Mohamed Mohandes
Keyword(s):  
Deep Learning ◽  
Sign Language ◽  
Markov Models ◽  
Body Language ◽  
Arab Countries ◽  
Microsoft Kinect ◽  
Classification Model ◽  
Language Recognition ◽  
Sign Language Recognition ◽  
Arabic Sign Language

Sign language is the major means of communication for the deaf community. It uses body language and gestures such as hand shapes, lib patterns, and facial expressions to convey a message. Sign language is geography-specific, as it differs from one country to another. Arabic Sign language is used in all Arab countries. The availability of a comprehensive benchmarking database for ArSL is one of the challenges of the automatic recognition of Arabic Sign language. This article introduces KArSL database for ArSL, consisting of 502 signs that cover 11 chapters of ArSL dictionary. Signs in KArSL database are performed by three professional signers, and each sign is repeated 50 times by each signer. The database is recorded using state-of-art multi-modal Microsoft Kinect V2. We also propose three approaches for sign language recognition using this database. The proposed systems are Hidden Markov Models, deep learning images’ classification model applied on an image composed of shots of the video of the sign, and attention-based deep learning captioning system. Recognition accuracies of these systems indicate their suitability for such a large number of Arabic signs. The techniques are also tested on a publicly available database. KArSL database will be made freely available for interested researchers.

scholarly journals Image Classification for the Automatic Feature Extraction in Human Worn Fashion Data

Mathematics ◽  
2021 ◽  
Vol 9 (6) ◽  
pp. 624
Author(s):  
Stefan Rohrmanstorfer ◽  
Mikhail Komarov ◽  
Felix Mödritscher
Keyword(s):  
Neural Networks ◽  
Feature Extraction ◽  
Image Classification ◽  
Convolutional Neural Networks ◽  
Data Augmentation ◽  
State Of The Art ◽  
Image Data ◽  
Classification Model ◽  
Upper Body ◽  
Automatic Feature Extraction

With the always increasing amount of image data, it has become a necessity to automatically look for and process information in these images. As fashion is captured in images, the fashion sector provides the perfect foundation to be supported by the integration of a service or application that is built on an image classification model. In this article, the state of the art for image classification is analyzed and discussed. Based on the elaborated knowledge, four different approaches will be implemented to successfully extract features out of fashion data. For this purpose, a human-worn fashion dataset with 2567 images was created, but it was significantly enlarged by the performed image operations. The results show that convolutional neural networks are the undisputed standard for classifying images, and that TensorFlow is the best library to build them. Moreover, through the introduction of dropout layers, data augmentation and transfer learning, model overfitting was successfully prevented, and it was possible to incrementally improve the validation accuracy of the created dataset from an initial 69% to a final validation accuracy of 84%. More distinct apparel like trousers, shoes and hats were better classified than other upper body clothes.

A Novel ECG Automatic Detection Using LongShort-Term Memory Network and Internet of Things Technology

2021 ◽  
Vol 11 (6) ◽  
pp. 1592-1598
Author(s):  
Xufei Liu
Keyword(s):  
Machine Learning ◽  
Cardiovascular Diseases ◽  
Internet Of Things ◽  
Network Structure ◽  
Recognition Rate ◽  
Classification Model ◽  
Data Set ◽  
Public Data ◽  
Lstm Network ◽  
Internet Of Things Technology

The early detection of cardiovascular diseases based on electrocardiogram (ECG) is very important for the timely treatment of cardiovascular patients, which increases the survival rate of patients. ECG is a visual representation that describes changes in cardiac bioelectricity and is the basis for detecting heart health. With the rise of edge machine learning and Internet of Things (IoT) technologies, small machine learning models have received attention. This study proposes an ECG automatic classification method based on Internet of Things technology and LSTM network to achieve early monitoring and early prevention of cardiovascular diseases. Specifically, this paper first proposes a single-layer bidirectional LSTM network structure. Make full use of the timing-dependent features of the sampling points before and after to automatically extract features. The network structure is more lightweight and the calculation complexity is lower. In order to verify the effectiveness of the proposed classification model, the relevant comparison algorithm is used to verify on the MIT-BIH public data set. Secondly, the model is embedded in a wearable device to automatically classify the collected ECG. Finally, when an abnormality is detected, the user is alerted by an alarm. The experimental results show that the proposed model has a simple structure and a high classification and recognition rate, which can meet the needs of wearable devices for monitoring ECG of patients.

scholarly journals Competing for Recognition through Public Good Provision

2008 ◽  
Vol 8 (1) ◽  
Author(s):  
Mattias K Polborn
Keyword(s):  
Public Good ◽  
Open Source Software ◽  
Public Good Provision ◽  
Competition Effect ◽  
Software Projects ◽  
The Public ◽  
Context Recognition ◽  
Private Providers ◽  
Common Interests ◽  
Good Provision

We consider a setting in which several groups of individuals with common interests (``clubs") compete with each other for recognition by other individuals. Depending on the context, recognition may be expressed by these other individuals joining a club, or choosing one club to admire. Clubs compete by providing a public good. Competition between clubs increases the public good provision level, and a sufficiently strong competition effect may even lead to overprovision. The model thus limits the argument for subsidies to the private providers of public goods. We discuss implications of the model for open-source software projects, university fundraising and infrastructure competition between cities.

scholarly journals Portfolio management and appropriation of private benefits of control

2016 ◽  
Vol 1 (1) ◽  
pp. 20
Author(s):  
Elli Kraizberg
Keyword(s):  
Portfolio Management ◽  
Empirical Finding ◽  
Expected Returns ◽  
Minority Shareholders ◽  
Publicly Traded ◽  
Final Model ◽  
Public Firms ◽  
The Public ◽  
Portfolio Managers ◽  
The Impact

<p dir="LTR">In many countries around the globe, portfolio managers utilize well accepted models, assuming that a partial stake of ownership is proportionally valued. This assumption is incorrect  in markets in which traded firms or publicly held firms are controlled by major owners who would take any possible measure to protect and maintain a 'lock' on control, so they can secure a sellable asset to another control seeker. In this case, estimation of key parameters such as, volatility, expected returns and diversification effect, may be grossly distorted.</p><p dir="LTR">We would argue that a major trigger for the value of the benefits of control is the ability of control owners to transfer assets from their own portfolio to a controlled publicly traded firm. While it is obvious that these transfers will take place, if and only if, it is beneficial to the control owners, the impact on the minor shareholders may not necessarily be negative and may vary depending on several parameters. Thus, the benefits of control are not entirely "private", i.e. appropriation and diversion of the resources of publicly traded firms for the benefit of the control owners.     </p><p dir="LTR">This paper aims to model the effect of the benefits of control on the value of a minority held public firms. It focuses on two related issues that are discussed in the literature on the benefits of control: what drives the value of the benefits of control, given the   empirical evidence that control seekers are willing to pay a significant premium for control, and secondly, can these benefits be rationally modeled? To better understand these issues, it then models a specific drive on the part of control seekers who, in addition to their stake in a publicly traded firm, own a private portfolio. It could be argued that they may 'transfer' inferior investments to the public firms that they control exploiting less than perfect transparency. However, while they own this valuable option of 'transferring' inferior investments into the public firm, these actions may still be beneficial to the minority shareholders.</p><p dir="LTR">We establish a model and derive a simulation procedure that are applied to several cases in which transfers  are made in exchange for cash or equity, instances of full disclosure or partial transparency, the likelihood that the control owners' actions will be contested in court, level of risk, and other parameters. Then we will compare the results to empirical finding.  The final model will be greatly simplified so that the end formula can be easily used by practitioners. </p>

scholarly journals An open and reproducible analysis of a controversial criminal law decision: Mission IMM-possible?

2020 ◽  
Author(s):  
Jason Chin
Keyword(s):  
Open Source Software ◽  
Research Process ◽  
High Court ◽  
The Public ◽  
Judicial Decisions ◽  
Research Fields ◽  
Law And Policy ◽  
Criminological Research ◽  
Reproducible Analysis ◽  
High Court Of Australia

Reproducibility and open access are central to the research process, enabling researchers to verify and build upon each other’s work, and allowing the public to rely on that work. These ideals are perhaps even more important in legal and criminological research, fields that actively seek to inform law and policy. This article has two goals. First, it seeks to advance legal and criminological research methods by serving as an example of a reproducible and open analysis of a controversial criminal evidence decision. Towards that end, this study relies on open source software, and includes an app (https://openlaw.shinyapps.io/imm-app/) allowing readers to access and read through the judicial decisions being analysed. The second goal is to examine the effect of the 2016 High Court of Australia decision, IMM v The Queen, which appeared to limit safeguards against evidence known to contribute to wrongful convictions in Australia and abroad.

scholarly journals Securing the wireless emergency alerts system

2021 ◽  
Vol 64 (10) ◽  
pp. 85-93
Author(s):  
Jihoon Lee ◽  
Gyuhong Lee ◽  
Jinsung Lee ◽  
Youngbin Im ◽  
Max Hollingsworth ◽  
...  
Keyword(s):  
Open Source ◽  
Open Source Software ◽  
Software Defined Radio ◽  
Severe Weather ◽  
Base Stations ◽  
The Public ◽  
Software Libraries ◽  
Amber Alerts ◽  
Almost All ◽  
To Receive

Modern cell phones are required to receive and display alerts via the Wireless Emergency Alert (WEA) program, under the mandate of the Warning, Alert, and Response Act of 2006. These alerts include AMBER alerts, severe weather alerts, and (unblockable) Presidential Alerts, intended to inform the public of imminent threats. Recently, a test Presidential Alert was sent to all capable phones in the U.S., prompting concerns about how the underlying WEA protocol could be misused or attacked. In this paper, we investigate the details of this system and develop and demonstrate the first practical spoofing attack on Presidential Alerts, using commercially available hardware and modified open source software. Our attack can be performed using a commercially available software-defined radio, and our modifications to the open source software libraries. We find that with only four malicious portable base stations of a single Watt of transmit power each, almost all of a 50,000-seat stadium can be attacked with a 90% success rate. The real impact of such an attack would, of course, depend on the density of cellphones in range; fake alerts in crowded cities or stadiums could potentially result in cascades of panic. Fixing this problem will require a large collaborative effort between carriers, government stakeholders, and cellphone manufacturers. To seed this effort, we also propose three mitigation solutions to address this threat.

Sign in / Sign up

Export Citation Format

Share Document