Fast face detection using a unified architecture for unconstrained and infrared face images

Human face detection along with its localization is a difficult task when the face is presented in the cluttered scene in an unconstrained scenario that might be with arbitrary pose variations, occlusions, random backgrounds and infrared (IR) environment. This paper proposes a novel face detection method which can address some of these issues and challenges quite successfully during face detection in unconstrained as well as infrared environments. It makes use of Fast Successive Mean Quantization Transform (FastSMQT) features for image enhancement and feature representation to deal with illumination and sensor insensitiveness. A split up Sparse Network of Winnows (SNoW) with Winnow updating rule is then exploited to speed up the original SNoW classifier. Finally, the features and classifiers are combined together with skin detection algorithm for face detection in crowd image and head orientation correction of near infrared faces. The proposed face detector is robust in handling pose, occlusion, illumination, blur and low image resolution. The experiment is performed on six challenging and publicly available databases, viz. BIOID, LFW, FDDB, UFI, WIDER FACE and IIT Delhi near infrared. The experimental results depict that the proposed method outperforms traditional as well as some advanced methods in detecting unconstrained and infrared faces under challenging situations.     

Low-level orientation information for social evaluation in face images

Observers make a range of social evaluations based on facial appearance, including judgments of trustworthiness, warmth, competence, and other aspects of personality. What visual information do people use to make these judgments? While links have been made between perceived social characteristics and other high-level properties of facial appearance (e.g., attractiveness, masculinity), there has been comparatively little effort to link social evaluations to low-level visual features, like spatial frequency and orientation sub-bands, known to be critically important for face processing. We explored the extent to which different social evaluations depended critically on horizontal orientation energy vs. vertical orientation energy, as is the case for face identification and emotion recognition. We found that while trustworthiness judgments exhibited this bias for horizontal orientations, competence and dominance did not, suggesting that social evaluations may depend on a multi-channel representation of facial appearance at early stages of visual processing.     

An X Windows tool for synthesizing face images from eigenvectors

An X Windows software tool for the construction of faces with a weighted combination of eigenvectors is described. The eigenvectors were extracted from an autoassociative matrix that comprised 100 face images. The program input consists of eigenvectors and sets of weights that describe individual faces and combines these to create face images. The tool creates a panel of buttons that permits the display of individual eigenvectors and the display of an average face as well. Facilities for on-line changes to the intensity of individual eigenvectors can be used to change the appearance of a face. Previously, O’Toole, Abdi, Deffenbacher, and Bartlett (1991) have shown that the intensity of certain individual eigenvectors contains reliable information for determining the sex and race of the face.     

Mapping physical characteristics in face images to social judgements

Sutherland and Young (Br. J. Psychol., 113, 2022, 1056) provide a comprehensive and timely overview of recent developments in research on social judgements of faces, emphasizing the utility of data-driven approaches. Here, we expand on this theme, focusing on how data-driven approaches can provide new insights into the physical characteristics in face images that best predict social judgements.     

Biased lineup instructions and face identification from video images

Previous eyewitness memory research has shown that biased lineup instructions reduce identification accuracy, primarily by increasing false-positive identifications in target-absent lineups. Because some attempts at identification do not rely on a witness's memory of the perpetrator but instead involve matching photos to images on surveillance video, the authors investigated the effects of biased instructions on identification accuracy in a matching task. In Experiment 1, biased instructions did not affect the overall accuracy of participants who used video images as an identification aid, but nearly all correct decisions occurred with target-present photo spreads. Both biased and unbiased instructions resulted in high false-positive rates. In Experiment 2, which focused on video-photo matching accuracy with target-absent photo spreads, unbiased instructions led to more correct responses (i.e., fewer false positives). These findings suggest that investigators should not relax precautions against biased instructions when people attempt to match photos to an unfamiliar person recorded on video.     

Digital image-processing techniques for the display of images and modeling of visual perception

With the rapid increase in power of image-processing technology, applications in the areas of robot and human vision are inevitable. In this review, I consider how such techniques make it possible to study new dimensions of human vision and I suggest explanations of visual phenomena, which, until now, have not been considered in detail.     

A bottleneck in face identification: repetition priming from flanker images

There is evidence that face processing is capacity-limited in distractor interference tasks and in tasks requiring overt recognition memory. We examined whether capacity limits for faces can be observed with a more sensitive measure of visual processing, by measuring repetition priming of flanker faces that were presented alongside a face or a nonface target. In Experiment 1, we found identity priming for face flankers, by measuring repetition priming across a change in image, during task-relevant nonface processing, but not during the processing of a concurrently-presented face target. Experiment 2 showed perceptual priming of the flanker faces, across identical images at prime and test, when they were presented alongside a face target. In a third Experiment, all of these effects were replicated by measuring identity priming and perceptual priming within the same task. Overall, these results imply that face processing is capacity limited, such that only a single face can be identified at one time. Merely attending to a target face appears sufficient to trigger these capacity limits, thereby extinguishing identification of a second face in the display, although our results demonstrate that the additional face remains at least subject to superficial image processing.     

Automated classification of demographics from face images: A tutorial and validation

Examining disparities in social outcomes as a function of gender, age, or race has a long tradition in psychology and other social sciences. With an increasing availability of large naturalistic data sets, researchers are afforded the opportunity to study the effects of demographic characteristics with real‐world data and high statistical power. However, since traditional studies rely on human raters to asses demographic characteristics, limits in participant pools can hinder researchers from analyzing large data sets. Automated procedures offer a new solution to the classification of face images. Here, we present a tutorial on how to use two face classification algorithms, Face++ and Kairos. We also test and compare their accuracy under varying conditions and provide practical recommendations for their use. Drawing on two face databases (n = 2,805 images), we find that classification accuracy is (a) relatively high, with Kairos generally outperforming Face++ (b) similar for standardized and more variable images, and (c) dependent on target demographics. For example, accuracy was lower for Hispanic and Asian (vs. Black and White) targets. In sum, we propose that automated face classification can be a useful tool for researchers interested in studying the effects of demographic characteristics in large naturalistic data sets.     

Controlling Stroop effects by manipulating expectations for color words

An important characteristic of automatic processing is its uncontrollability. The Stroop phenomenon is regarded as a prototypical example of this characteristic of automatic processing, hence, the Stroop effect should not change when the percentages of color words versus neutral stimuli are manipulated to induce controlled processing. We found that Stroop interference decreased as the percentage of color words increased. Furthermore, the magnitude of the inhibitory component of the Stroop effect was negatively correlated with the percentage of color words; the facilitatory component was insensitive to the manipulation. These results suggest that the Stroop effect is controllable (see Logan, 1980) and that the locus of control is postlexical. The results also suggest that facilitation and inhibition are produced by different mechanisms and challenge those models of the Stroop phenomenon (e.g., Cohen, Dunbar, & McClelland, 1990; Phaf, Van der Heijden, & Hudson, 1990) that assume that a single processing mechanism causes facilitation and inhibition and that control affects facilitation and inhibition alike (Logan, 1980).     

ALICE: A system for manipulating and analyzing multidimensional data

The computer program ALICE solves the two major problems of data manipulation and analysis. First, ALICE allows the user to treat data from an experiment in the form they are generated. Second, mathematical calculations and statistical analyses are included as an intrinsic part of the multidimensional approach to data handling. ALICE accepts raw data in the form and order they were collected; reorganizes, partitions, or selects any subset of them (including a single entry), and arithmetically combines, transforms, or evaluates any formula involving them. Furthermore, learning to use ALICE is simple, even for those who are naive to both computers and data analysis.     

Identity-specific face adaptation effects: evidence for abstractive face representations

The effects of selective adaptation on familiar face perception were examined. After prolonged exposure to photographs of a celebrity, participants saw a series of ambiguous morphs that were varying mixtures between the face of that person and a different celebrity. Participants judged fewer of the morphs to resemble the celebrity to which they had been adapted, implying that they were now less sensitive to that particular face. Similar results were obtained when the adapting faces were highly dissimilar in viewpoint to the test morphs; when they were presented upside-down; or when they were vertically stretched to three times their normal height. These effects rule out explanations of adaptation effects solely in terms of low-level image-based adaptation. Instead they are consistent with the idea that relatively viewpoint-independent, person-specific adaptation occurred, at the level of either the “Face Recognition Units” or “Person Identity Nodes” in Burton, Bruce and Johnston’s (1990) model of face recognition.     

A simple procedure for accurately manipulating face-voice synchrony when dubbing speech onto videotape

A system is described involving an audio-video tape recorder, a control device, and a delay timer that dubs speech sounds onto prerecorded lip movements on videotape. Accuracy of synchrony or measured asynchrony between the sound and lip movements is a few milliseconds.     

A Single-Subject Evaluation of the Treatment of Morphing Fear

We present a single-subject prospective outcome study of a man with severe morphing fear and long history of OCD who was not helped by previous interventions, and who received an adapted form of cognitive behavior therapy (CBT) as part of this study. Treatment consisted of a cognitively focused approach tailored to address his fear of morphing and included developing a stronger sense of self-stability. We describe the details of the case, the treatment protocol, and the therapeutic outcomes as assessed over 36 weeks by questionnaires, rating scales, and semistructured interviews. The intervention was effective in eradicating the patient’s morphing fears and reducing other symptoms of OCD, anxiety, and depression. The presented case illustrates the need to appropriately conceptualize, assess, and address the specific nature of morphing fear symptoms in treatment.     

Concurrent performances: synthesizing rate constancies by manipulating contingencies for a single response

An earlier experiment scheduled variable-interval reinforcement for pigeons' pecks on one key, and variable-interval reinforcement alternating with extinction, in a multiple schedule, for pecks on a second key. During the second key's extinction component, first-key pecking was relatively slow and continuous, rarely interrupted by second-key pecking; during the variable-interval component, first-key pecking was frequently interrupted by second-key pecking. When changeover delays operated, so that reinforced pecks on one key could not follow closely upon changeovers from the other key, rapid first-key pecking between interruptions compensated sufficiently for the time lost in second-key pecking that the overall rate of first-key pecking remained roughly constant across the alternating multiple-schedule components. The present experiments duplicated, on a single key, the temporal pattern of first-key pecking generated in the earlier experiments: components of continuous key availability were alternated with components of interrupted key availability. Approximately constant overall rates of responding were observed with a single-key equivalent of a changeover delay scheduled after interruptions and with manipulations of the on-off durations of the interruption cycle. Rate constancies in the original concurrent situation presumably depended on analogous contingencies that operated upon the concurrent responses, rather than on any constant “reserve” of responses.