Search Results

This chapter presents an approach for developing a fully automatic Facial Action Coding System (FACS). The approach uses state-of-the-art machine learning techniques that can be applied to recognition of any facial action. The results of Study I provided guidance as to which image representations, or feature extraction methods, are most effective for facial action recognition. Gabor wavelets and Independent Component Analysis gave best performance. Study II found that machine learning techniques applied directly to the warped images is a promising approach for automatic coding of spontaneous facial expressions. Generally, the data employed hand-labeled feature points for the head pose tracking step. Furthermore, three of the issues are discussed in detail: (1) collection of a database of spontaneous facial expressions, (2) fully automatic face detection and tracking, and (3) fully automatic 3D head pose estimation.

In this chapter we define the problem space and describe the core components of automatic facial expression recognition systems. In particular, we discuss the most prominent methods of face segmentation, face registration, feature extraction, classification, and temporal integration. We then present several practical applications of expression recognition technology. Finally, we comment on the core future challenges to the field, including generalization to multiple poses and ethnicities, collection of better training data, evaluation infrastructure, and commercialization.

This chapter examines how automated facial recognition and related technologies were envisioned and designed to serve a set of institutional priorities during the period of political-economic neoliberalization in the United States. In the United States the effort to program computers to identify human faces began in the 1960s in research labs funded by the Department of Defense and intelligence agencies. By the 1990s new companies were formed to commercialize the technology, searching for market especially among institutions operating proprietary computer networks and large-scale identification systems. Across these sectors, biometric identification promised to enable the “securitization of identity,” the intensification of identification practices at a proliferation of sites—a priority that has gone hand in hand with political-economic and governmental neoliberalization.

This chapter explores the development of automated facial expression analysis, the effort to program computers to recognize facial expressions as they form on and move across people's faces. While facial recognition technology treats the face as a “blank somatic surface” to be differentiated from other faces as an index of identity, automated facial expression analysis treats the dynamic surface of the face as the site of differentiation. In automated facial expression analysis, the dimensions and intensities of facial movements are analyzed as indices of emotion and cognition to determine what people are thinking and feeling. Automation of facial expression analysis promises to accomplish what facial recognition technology fails to do: read the interior of the person off the surface of the face, using the face itself as a field of classifiable information about the individual.

This introductory chapter provides an overview of facial recognition technology (FRT). In the post 9/11 context, FRT emerged as an existing, reliable, and high-tech solution to the most pressing problem facing the United States. Indeed, new forms of human-machine integration promise to make surveillance systems function more effectively and extend their reach over time and space. However, whether these experimental technologies can or should be made to accomplish these goals remains open to debate, one that often plays out in press and policy discussion as a trade-off between “security” and “privacy.” This book explores the web of cultural practices and ideas, along with the policy programs and institutional priorities, in which automated face perception technologies are embedded.

This chapter discusses the idea of “Smart CCTV”—the integration of automated facial recognition with video surveillance. Recognizing a business opportunity, entrepreneurs attempting to commercialize facial recognition technology introduced the idea as a potential solution to the problems of surveillance labor and video overload. In June 2001, Visionics Corporation began a project with the Tampa Police Department (TPD) to incorporate the company's automated facial recognition product, called FaceIt, into Ybor City. However, after a two-year free trial period, the Tampa Police abandoned the effort to integrate automated facial recognition with the Ybor City CCTV system, citing its failure to identify a single wanted individual. The failure of this experiment demonstrates that building a functioning Smart CCTV system involves resolving a complex combination of social and technological problems.

Neuropsychological evidence suggests that the recognition of facial identity is mediated by a distributed neural system. Damage to the temporal versus frontal lobe components of this face memory network produces qualitatively different types of recognition impairments. Specifically, whereas temporal lobe lesions result in clinical syndromes characterized by memory loss for familiar faces (prosopagnosia, person recognition disorders), frontal lobe lesions give rise to memory distortions involving false recognition/misidentification of unfamiliar faces. To account for these observations, we propose that dynamic functional interactions between temporal lobe memory and frontal executive systems normally play a critical role in maintaining face recognition accuracy by modulating the relationship between two complementary cognitive operations that involve processing faces at different levels of specificity: individuation and categorization.

James Simpson’s central hermeneutic perception for knowledge in the Humanities is that cognition is re-cognition. Before we can know, we must already have known. He examines this paradox with reference to literary examples of facial recognition from, in particular, Chaucer and his reception in the early modern period. Linking literary face to textual face – the whole text as a kind of face – he applies the lessons learnt from facial recognition to textual recognition; and answers some possible objections to the paradox of knowing being dependent on having already known.

This chapter examines how social uses of facial recognition and other biometrics are being defined at the level of individual users, and how the securitization of identity is incorporated at the level of individual practice. While biometric identification systems are being envisioned and designed mainly for large-scale institutional uses, developers have experimented with consumer applications. These consumer applications encourage the cooperation of individual users in the process of biometric system development and institutionalization. Moreover, consumer application of biometrics are designed with the new responsibilities and competencies of the security-conscious, tech-savvy citizen in mind. The chapter then looks at two types of consumer applications: personal security applications and facial recognition software designed for the management of personal photo collections and online photo sharing.

This chapter studies the preoccupation with facial recognition technology (FRT) in the post-9/11 context. It considers claims of its technical neutrality by investigating the cultural logic that defined the technology and the practical politics that shaped system development. While the promise of facial recognition lay in its potential to individualize the terrorist threat by targeting specifically identified “terrorist” individuals, the effort to define it as a homeland security technology also made use of an implicit classifying logic, including rhetorical moves that reinforced old-fashioned notions of deviant facial types. Thus, the “facialization” of terrorism—defining non-state forms of political violence with recourse to the racist logic of a mythic and demonized facial type—was prevalent in discourse about FRT, appearing alongside claims about its technical neutrality.

This book discusses developments that have occurred recently in the field of the recognition and modeling of dynamic faces. It discusses the topics in neuroscience that are important for researchers in biological science, neuroscience, and computer science, as well as the psychophysics of dynamic face perception and computational models depicting how dynamic faces are processed in the brain. The book emphasizes that the principles used by the brain to recognize dynamic faces help to provide technical solutions in computer vision. Additionally, knowledge required for improving technical systems for the recognition and animation of dynamic faces has significance in the field of computer science.

This chapter discusses research on facial self-recognition. The ability to recognize one’s own face, for instance by passing the mirror mark test, has often been heralded as providing empirical evidence for the presence of self-consciousness. A failure to pass the test has also been seen as evidence for the absence of self-consciousness. Some, such as Gallup, have even argued that creatures incapable of passing such a test lack conscious experiences altogether. The latter interpretations are criticized, and the plausibility of an alternative interpretation of mirror self-experience is assessed, one that sees facial self-recognition as testifying to the presence of a rather special kind of self-consciousness, namely, one that in the case of human beings often has a distinctive social dimension to it.

Life-altering technology is not only improving our lifestyle but also creating new business models. Integration of technology into everyday life is a primary driver of changes in lifestyle. Whether visible or not, today’s technology is everywhere. Consumers come home from work to a smart house that greets them with music, emails them the foods the refrigerator needs, and through spatial phase imaging technology senses their mood. Without human intervention it changes its presentation based on data indicators embedded in everything. The house recognizes mood and compares it with past behaviors, facial reactions, timeline, and acts accordingly. All this happens through a standard security camera with pixelate three-dimensional technology. The same technology can identify the anti-social elements in a crowd, enhance security at any public event venue, and allow doctors to see under our skin without intrusion.

Uncanny aesthetics examines how the discourse of the uncanny emerged from a reading of E. T. A Hoffmann’s short story “The Sandman,” in the works of Ernst Jentsch and Sigmund Freud, and how these readings have shaped the way we think about the uncanny as both experiential and aesthetic. These interpretations demonstrate, as Samuel Weber puts it, “a certain indecidability” between what we personally experience and what is predetermined. The uncanny has, however, shifted from a fear of confronting unhuman objects to the fear of being exposed to others beyond the devices and programs with which we have intimate relations, and the existential crisis of not measuring up to the technologies that simulate us.

Chapter 2 presents the divergent data sets, collected in 2014, 2015, 2016, and 2017 and focusing primarily on the 2014 U.S. Senate campaigns, used to test the tolerance and tactics theory of negativity. U.S. Senate elections are the ideal laboratory for exploring the impact of negative advertising because senatorial campaigns are characterized by impressive variability in the amount, content, and tone of negativity, the types of candidates, the size of the constituencies, and the characteristics of the media markets. A variety of methods, including surveys, experiments, content analyses, focus groups, and facial recognition software are used to measure people’s tolerance for negativity; the relevance and civility of negative advertisements; and people’s reactions to negative commercials varying in civility and relevance. These various data sets are ultimately used to assess the impact of negative advertisements on people’s assessments of candidates and on their decision to vote on Election Day.