Simulation of Border Control in an Ongoing Web-based Experiment for Estimating Morphing Detection Performance of Humans

Makrushin, Andrey; Siegel, Dennis; Dittmann, Jana

doi:10.1145/3369412.3395073

Cited by 12 publications

(19 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A similar web-based experiment simulating border control was presented by Makrushin et al [74], who studied human perception analysis by both skilled and unskilled humans and further extended [75] to obtain more unbiased and realistic images. In both cases, skilled humans (who have knowledge of morphed face images) show the best performance in detecting a morphed face image.…”

Section: Human Perception and Morphed Face Detectionmentioning

confidence: 97%

“…Thus, the border guard considers the facial similarity of the traveller to the reference data in the passport to make his/her final decision. Several studies in the literature have indicated the effectiveness of morphed images in deceiving expert human observers [49], [60], [65], [73]- [75], [98], [99], [127]. Early investigations on human perception analysis of morphed images were reported by Jäger et al [60], where different experiments were performed to benchmark the ability of human observers to detect face morphing and its dependency on various parameters (i.e., different alpha/morphing factors).…”

Section: Human Perception and Morphed Face Detectionmentioning

confidence: 99%

“…Summarising the works on human perception analysis, it is noted that both skilled and unskilled human observers often fail to detect morphed face images. However, it is also noted that considerable training of human observers can improve morphed face detection [75], [99].…”

Section: Human Perception and Morphed Face Detectionmentioning

confidence: 99%

See 2 more Smart Citations

Face Morphing Attack Generation and Detection: A Comprehensive Survey

Venkatesh

Ramachandra

Raja

et al. 2021

IEEE Trans. Technol. Soc.

100

View full text Add to dashboard Cite

Face recognition has been successfully deployed in real-time applications, including secure applications such as border control. The vulnerability of face recognition systems (FRSs) to various kinds of attacks (both direct and indirect attacks) and face morphing attacks has received great interest from the biometric community. The goal of a morphing attack is to subvert an FRS at an automatic border control (ABC) gate by presenting an electronic machine-readable travel document (eMRTD) or e-passport that is obtained based on a morphed face image. Since the application process for an e-passport in the majority of countries requires a passport photo to be presented by the applicant, a malicious actor and an accomplice can generate a morphed face image to obtain the e-passport. An e-passport with a morphed face image can be used by both the malicious actor and the accomplice to cross a border, as the morphed face image can be verified against both of them. This can result in a significant threat, as a malicious actor can cross the border without revealing the trace of his/her criminal background, while the details of the accomplice are recorded in the log of the access control system. This survey aims to present a systematic overview of the progress made in the area of face morphing in terms of both morph generation and morph detection. In this paper, we describe and illustrate various aspects of face morphing attacks, including different techniques for generating morphed face images and state-of-the-art morph attack detection (MAD) algorithms based on a stringent taxonomy as well as the availability of public databases, which allow us to benchmark new MAD algorithms in a reproducible manner. The outcomes of competitions and benchmarking, vulnerability assessments and performance evaluation metrics are also provided in a comprehensive manner. Furthermore, we discuss the open challenges and potential future areas that need to be addressed in the evolving field of biometrics.

show abstract

Section: Human Perception and Morphed Face Detectionmentioning

confidence: 97%

Section: Human Perception and Morphed Face Detectionmentioning

confidence: 99%

See 1 more Smart Citation

Face Morphing Attack Generation and Detection: A Comprehensive Survey

Venkatesh

Ramachandra

Raja

et al. 2021

IEEE Trans. Technol. Soc.

100

View full text Add to dashboard Cite

show abstract

“…This is particularly true for still images, which today's GANs impressively synthesise through easy-to-use interfaces. 7 By contrast, the creation of realistic manipulated videos of arbitrary subjects and in high resolution still requires significant resources and expertise. We indicate them as AI-17, AI-18, and AI-19 and keep this convention throughout the column.…”

mentioning

confidence: 99%

“…6 https://edition.cnn.com/2020/02/28/tech/fake-twitter-candidate-2020/index.html 7 See for instance https://thispersondoesnotexist.com/…”

mentioning

confidence: 99%

More Real than Real: A Study on Human Visual Perception of Synthetic Faces

Lago,

Pasquini,

Böhme

et al. 2021

Preprint

View full text Add to dashboard Cite

Deep fakes have become popular recently. The term refers to doctored media contents where one face's is swapped with someone else's face or performs someone else's face movements. In the last couple of years, numerous video clips, often involving celebrities and politicians, have gone viral on social media platforms. This has been enabled by easy to use apps capable to process user-generated content in real-time.Informally, deep fakes can be defined as realistic digital media (images, videos, or audio tracks) depicting untruthful content, obtained either by manipulating pristine material or generated from scratch. The attribute deep refers to the use of algorithms based on deep learning, a subfield of modern Artificial Intelligence (AI), which pushed the boundaries for many applications including media data manipulation and generation. Besides offering exciting opportunities in several fields (such as entertainment, content production, e-learning, and e-health), these advanced creation technologies are now widely recognized as a pressing threat to the reliability of visual information. 1This may have severe consequences for the digital identity and reputation of individuals.Many cases of misuse reported in the past months demonstrate the potential impact of manipulated data on disinformation. For example, the last US presidential campaign witnessed the viral diffusion of multiple deceptive visual manipulations: amongst others, an altered video of Joe Biden greeting the wrong state in a public speech, as well as retouched pictures of celebrities untruthfully implying their endorsement for Donald Trump. 2 Not very surprisingly, deep fakes become increasingly ubiquitous on the web: the estimated number of deep fakes on the web doubled every six months since 2018, reaching a total of 85'000 in 2020, 3 although it is not possible to tell which 1 https://www.theguardian.com/technology/ng-interactive/2019/jun/22/the-rise-of-thedeepfake-and-the-threat-to-democracy 2 https://www.fastcompany.com/90575763/we-have-the-technology-to-fight-manipulatedimages-and-videos-its-time-to-use-it 3 https://sensity.ai/how-to-detect-a-deepfake-with-sensity/

show abstract

Psychophysical Evaluation of Human Performance in Detecting Digital Face Image Manipulations

et al. 2022

View full text Add to dashboard Cite

In recent years, increasing deployment of face recognition technology in security-critical settings, such as border control or law enforcement, has led to considerable interest in the vulnerability of face recognition systems to attacks utilising legitimate documents, which are issued on the basis of digitally manipulated face images. As automated manipulation and attack detection remains a challenging task, conventional processes with human inspectors performing identity verification remain indispensable. These circumstances merit a closer investigation of human capabilities in detecting manipulated face images, as previous work in this field is sparse and often concentrated only on specific scenarios and biometric characteristics. This work introduces a web-based, remote visual discrimination experiment on the basis of principles adopted from the field of psychophysics and subsequently discusses interdisciplinary opportunities with the aim of examining human proficiency in detecting different types of digitally manipulated face images, specifically face swapping, morphing, and retouching. In addition to analysing appropriate performance measures, a possible metric of detectability is explored. Experimental data of 306 probands indicate that detection performance is widely distributed across the population and detection of certain types of face image manipulations is much more challenging than others.

show abstract

Simulation of Border Control in an Ongoing Web-based Experiment for Estimating Morphing Detection Performance of Humans

Cited by 12 publications

References 16 publications

Face Morphing Attack Generation and Detection: A Comprehensive Survey

Face Morphing Attack Generation and Detection: A Comprehensive Survey

More Real than Real: A Study on Human Visual Perception of Synthetic Faces

Psychophysical Evaluation of Human Performance in Detecting Digital Face Image Manipulations

Contact Info

Product

Resources

About