Kevin W. Bowyer

Haiyu Wu, Vítor Albiero, K. S. Krishnapriya et al.

We explore varying face recognition accuracy across demographic groups as a phenomenon partly caused by differences in face illumination. We observe that for a common operational scenario with controlled image acquisition, there is a large difference in face region brightness between African-American and Caucasian, and also a smaller difference between male and female. We show that impostor image pairs with both faces under-exposed, or both overexposed, have an increased false match rate (FMR). Conversely, image pairs with strongly different face brightness have a decreased similarity measure. We propose a brightness information metric to measure variation in brightness in the face and show that face brightness that is too low or too high has reduced information in the face region, providing a cause for the lower accuracy. Based on this, for operational scenarios with controlled image acquisition, illumination should be adjusted for each individual to obtain appropriate face image brightness. This is the first work that we are aware of to explore how the level of brightness of the skin region in a pair of face images (rather than a single image) impacts face recognition accuracy, and to evaluate this as a systematic factor causing unequal accuracy across demographics. The code is at https://github.com/HaiyuWu/FaceBrightness.

Aidan Boyd, Jeremy Speth, Lucas Parzianello et al.

Research in presentation attack detection (PAD) for iris recognition has largely moved beyond evaluation in "closed-set" scenarios, to emphasize ability to generalize to presentation attack types not present in the training data. This paper offers several contributions to understand and extend the state-of-the-art in open-set iris PAD. First, it describes the most authoritative evaluation to date of iris PAD. We have curated the largest publicly-available image dataset for this problem, drawing from 26 benchmarks previously released by various groups, and adding 150,000 images being released with the journal version of this paper, to create a set of 450,000 images representing authentic iris and seven types of presentation attack instrument (PAI). We formulate a leave-one-PAI-out evaluation protocol, and show that even the best algorithms in the closed-set evaluations exhibit catastrophic failures on multiple attack types in the open-set scenario. This includes algorithms performing well in the most recent LivDet-Iris 2020 competition, which may come from the fact that the LivDet-Iris protocol emphasizes sequestered images rather than unseen attack types. Second, we evaluate the accuracy of five open-source iris presentation attack algorithms available today, one of which is newly-proposed in this paper, and build an ensemble method that beats the winner of the LivDet-Iris 2020 by a substantial margin. This paper demonstrates that closed-set iris PAD, when all PAIs are known during training, is a solved problem, with multiple algorithms showing very high accuracy, while open-set iris PAD, when evaluated correctly, is far from being solved. The newly-created dataset, new open-source algorithms, and evaluation protocol, made publicly available with the journal version of this paper, provide the experimental artifacts that researchers can use to measure progress on this important problem.

Rasel Ahmed Bhuiyan, Parisa Farmanifard, Renu Sharma et al.

Post-mortem iris recognition brings both hope to the forensic community (a short-term but accurate and fast means of verifying identity) as well as concerns to society (its potential illicit use in post-mortem impersonation). These hopes and concerns have grown along with the volume of research in post-mortem iris recognition. Barriers to further progress in post-mortem iris recognition include the difficult nature of data collection, and the resulting small number of approaches designed specifically for comparing iris images of deceased subjects. This paper makes several unique contributions to mitigate these barriers. First, we have collected and we offer a new dataset of NIR (compliant with ISO/IEC 19794-6 where possible) and visible-light iris images collected after demise from 259 subjects, with the largest PMI (post-mortem interval) being 1,674 hours. For one subject, the data has been collected before and after death, the first such case ever published. Second, the collected dataset was combined with publicly-available post-mortem samples to assess the current state of the art in automatic forensic iris recognition with five iris recognition methods and data originating from 338 deceased subjects. These experiments include analyses of how selected demographic factors influence recognition performance. Thirdly, this study implements a model for detecting post-mortem iris images, which can be considered as presentation attacks. Finally, we offer an open-source forensic tool integrating three post-mortem iris recognition methods with explainability elements added to make the comparison process more human-interpretable.

Aman Bhatta, Vítor Albiero, Kevin W. Bowyer et al.

It is broadly accepted that there is a "gender gap" in face recognition accuracy, with females having higher false match and false non-match rates. However, relatively little is known about the cause(s) of this gender gap. Even the recent NIST report on demographic effects lists "analyze cause and effect" under "what we did not do". We first demonstrate that female and male hairstyles have important differences that impact face recognition accuracy. In particular, compared to females, male facial hair contributes to creating a greater average difference in appearance between different male faces. We then demonstrate that when the data used to estimate recognition accuracy is balanced across gender for how hairstyles occlude the face, the initially observed gender gap in accuracy largely disappears. We show this result for two different matchers, and analyzing images of Caucasians and of African-Americans. These results suggest that future research on demographic variation in accuracy should include a check for balanced quality of the test data as part of the problem formulation. To promote reproducible research, matchers, attribute classifiers, and datasets used in this research are/will be publicly available.

Haiyu Wu, Grace Bezold, Aman Bhatta et al.

Face attribute research has so far used only simple binary attributes for facial hair; e.g., beard / no beard. We have created a new, more descriptive facial hair annotation scheme and applied it to create a new facial hair attribute dataset, FH37K. Face attribute research also so far has not dealt with logical consistency and completeness. For example, in prior research, an image might be classified as both having no beard and also having a goatee (a type of beard). We show that the test accuracy of previous classification methods on facial hair attribute classification drops significantly if logical consistency of classifications is enforced. We propose a logically consistent prediction loss, LCPLoss, to aid learning of logical consistency across attributes, and also a label compensation training strategy to eliminate the problem of no positive prediction across a set of related attributes. Using an attribute classifier trained on FH37K, we investigate how facial hair affects face recognition accuracy, including variation across demographics. Results show that similarity and difference in facial hairstyle have important effects on the impostor and genuine score distributions in face recognition. The code is at https:// github.com/ HaiyuWu/ LogicalConsistency.

Haiyu Wu, Kevin W. Bowyer

The issue of demographic disparities in face recognition accuracy has attracted increasing attention in recent years. Various face image datasets have been proposed as 'fair' or 'balanced' to assess the accuracy of face recognition algorithms across demographics. These datasets typically balance the number of identities and images across demographics. It is important to note that the number of identities and images in an evaluation dataset are {\em not} driving factors for 1-to-1 face matching accuracy. Moreover, balancing the number of identities and images does not ensure balance in other factors known to impact accuracy, such as head pose, brightness, and image quality. We demonstrate these issues using several recently proposed datasets. To improve the ability to perform less biased evaluations, we propose a bias-aware toolkit that facilitates creation of cross-demographic evaluation datasets balanced on factors mentioned in this paper.

Aidan Boyd, Patrick Tinsley, Kevin Bowyer et al.

Face image synthesis has progressed beyond the point at which humans can effectively distinguish authentic faces from synthetically generated ones. Recently developed synthetic face image detectors boast "better-than-human" discriminative ability, especially those guided by human perceptual intelligence during the model's training process. In this paper, we investigate whether these human-guided synthetic face detectors can assist non-expert human operators in the task of synthetic image detection when compared to models trained without human-guidance. We conducted a large-scale experiment with more than 1,560 subjects classifying whether an image shows an authentic or synthetically-generated face, and annotate regions that supported their decisions. In total, 56,015 annotations across 3,780 unique face images were collected. All subjects first examined samples without any AI support, followed by samples given (a) the AI's decision ("synthetic" or "authentic"), (b) class activation maps illustrating where the model deems salient for its decision, or (c) both the AI's decision and AI's saliency map. Synthetic faces were generated with six modern Generative Adversarial Networks. Interesting observations from this experiment include: (1) models trained with human-guidance offer better support to human examination of face images when compared to models trained traditionally using cross-entropy loss, (2) binary decisions presented to humans offers better support than saliency maps, (3) understanding the AI's accuracy helps humans to increase trust in a given model and thus increase their overall accuracy. This work demonstrates that although humans supported by machines achieve better-than-random accuracy of synthetic face detection, the ways of supplying humans with AI support and of building trust are key factors determining high effectiveness of the human-AI tandem.

Aidan Boyd, Daniel Moreira, Andrey Kuehlkamp et al.

Forensic iris recognition, as opposed to live iris recognition, is an emerging research area that leverages the discriminative power of iris biometrics to aid human examiners in their efforts to identify deceased persons. As a machine learning-based technique in a predominantly human-controlled task, forensic recognition serves as "back-up" to human expertise in the task of post-mortem identification. As such, the machine learning model must be (a) interpretable, and (b) post-mortem-specific, to account for changes in decaying eye tissue. In this work, we propose a method that satisfies both requirements, and that approaches the creation of a post-mortem-specific feature extractor in a novel way employing human perception. We first train a deep learning-based feature detector on post-mortem iris images, using annotations of image regions highlighted by humans as salient for their decision making. In effect, the method learns interpretable features directly from humans, rather than purely data-driven features. Second, regional iris codes (again, with human-driven filtering kernels) are used to pair detected iris patches, which are translated into pairwise, patch-based comparison scores. In this way, our method presents human examiners with human-understandable visual cues in order to justify the identification decision and corresponding confidence score. When tested on a dataset of post-mortem iris images collected from 259 deceased subjects, the proposed method places among the three best iris matchers, demonstrating better results than the commercial (non-human-interpretable) VeriEye approach. We propose a unique post-mortem iris recognition method trained with human saliency to give fully-interpretable comparison outcomes for use in the context of forensic examination, achieving state-of-the-art recognition performance.

Aman Bhatta, Gabriella Pangelinan, Michael C. King et al.

Most studies to date that have examined demographic variations in face recognition accuracy have analyzed 1-to-1 matching accuracy, using images that could be described as "government ID quality". This paper analyzes the accuracy of 1-to-many facial identification across demographic groups, and in the presence of blur and reduced resolution in the probe image as might occur in "surveillance camera quality" images. Cumulative match characteristic curves (CMC) are not appropriate for comparing propensity for rank-one recognition errors across demographics, and so we use three metrics for our analysis: (1) the well-known d' metric between mated and non-mated score distributions, and introduced in this work, (2) absolute score difference between thresholds in the high-similarity tail of the non-mated and the low-similarity tail of the mated distribution, and (3) distribution of (mated - non-mated rank-one scores) across the set of probe images. We find that demographic variation in 1-to-many accuracy does not entirely follow what has been observed in 1-to-1 matching accuracy. Also, different from 1-to-1 accuracy, demographic comparison of 1-to-many accuracy can be affected by different numbers of identities and images across demographics. More importantly, we show that increased blur in the probe image, or reduced resolution of the face in the probe image, can significantly increase the false positive identification rate. And we show that the demographic variation in these high blur or low resolution conditions is much larger for male / female than for African-American / Caucasian. The point that 1-to-many accuracy can potentially collapse in the context of processing "surveillance camera quality" probe images against a "government ID quality" gallery is an important one.

Colton Crum, Patrick Tinsley, Aidan Boyd et al.

The performance of convolutional neural networks has continued to improve over the last decade. At the same time, as model complexity grows, it becomes increasingly more difficult to explain model decisions. Such explanations may be of critical importance for reliable operation of human-machine pairing setups, or for model selection when the "best" model among many equally-accurate models must be established. Saliency maps represent one popular way of explaining model decisions by highlighting image regions models deem important when making a prediction. However, examining salience maps at scale is not practical. In this paper, we propose five novel methods of leveraging model salience to explain a model behavior at scale. These methods ask: (a) what is the average entropy for a model's salience maps, (b) how does model salience change when fed out-of-set samples, (c) how closely does model salience follow geometrical transformations, (d) what is the stability of model salience across independent training runs, and (e) how does model salience react to salience-guided image degradations. To assess the proposed measures on a concrete and topical problem, we conducted a series of experiments for the task of synthetic face detection with two types of models: those trained traditionally with cross-entropy loss, and those guided by human salience when training to increase model generalizability. These two types of models are characterized by different, interpretable properties of their salience maps, which allows for the evaluation of the correctness of the proposed measures. We offer source codes for each measure along with this paper.

Gabriella Pangelinan, K. S. Krishnapriya, Vitor Albiero et al.

In recent years, media reports have called out bias and racism in face recognition technology. We review experimental results exploring several speculated causes for asymmetric cross-demographic performance. We consider accuracy differences as represented by variations in non-mated (impostor) and / or mated (genuine) distributions for 1-to-1 face matching. Possible causes explored include differences in skin tone, face size and shape, imbalance in number of identities and images in the training data, and amount of face visible in the test data ("face pixels"). We find that demographic differences in face pixel information of the test images appear to most directly impact the resultant differences in face recognition accuracy.

Kagan Ozturk, Grace Bezold, Aman Bhatta et al.

A person's facial hairstyle, such as presence and size of beard, can significantly impact face recognition accuracy. There are publicly-available deep networks that achieve reasonable accuracy at binary attribute classification, such as beard / no beard, but few if any that segment the facial hair region. To investigate the effect of facial hair in a rigorous manner, we first created a set of fine-grained facial hair annotations to train a segmentation model and evaluate its accuracy across African-American and Caucasian face images. We then use our facial hair segmentations to categorize image pairs according to the degree of difference or similarity in the facial hairstyle. We find that the False Match Rate (FMR) for image pairs with different categories of facial hairstyle varies by a factor of over 10 for African-American males and over 25 for Caucasian males. To reduce the bias across image pairs with different facial hairstyles, we propose a scheme for adaptive thresholding based on facial hairstyle similarity. Evaluation on a subject-disjoint set of images shows that adaptive similarity thresholding based on facial hairstyles of the image pair reduces the ratio between the highest and lowest FMR across facial hairstyle categories for African-American from 10.7 to 1.8 and for Caucasians from 25.9 to 1.3. Facial hair annotations and facial hair segmentation model will be publicly available.

Colton R. Crum, Aidan Boyd, Kevin Bowyer et al.

Machine learning models have shown increased accuracy in classification tasks when the training process incorporates human perceptual information. However, a challenge in training human-guided models is the cost associated with collecting image annotations for human salience. Collecting annotation data for all images in a large training set can be prohibitively expensive. In this work, we utilize "teacher" models (trained on a small amount of human-annotated data) to annotate additional data by means of teacher models' saliency maps. Then, "student" models are trained using the larger amount of annotated training data. This approach makes it possible to supplement a limited number of human-supplied annotations with an arbitrarily large number of model-generated image annotations. We compare the accuracy achieved by our teacher-student training paradigm with (1) training using all available human salience annotations, and (2) using all available training data without human salience annotations. We use synthetic face detection and fake iris detection as example challenging problems, and report results across four model architectures (DenseNet, ResNet, Xception, and Inception), and two saliency estimation methods (CAM and RISE). Results show that our teacher-student training paradigm results in models that significantly exceed the performance of both baselines, demonstrating that our approach can usefully leverage a small amount of human annotations to generate salience maps for an arbitrary amount of additional training data.

Haiyu Wu, Grace Bezold, Manuel Günther et al.

We report the first systematic analysis of the experimental foundations of facial attribute classification. Two annotators independently assigning attribute values shows that only 12 of 40 common attributes are assigned values with >= 95% consistency, and three (high cheekbones, pointed nose, oval face) have essentially random consistency. Of 5,068 duplicate face appearances in CelebA, attributes have contradicting values on from 10 to 860 of the 5,068 duplicates. Manual audit of a subset of CelebA estimates error rates as high as 40% for (no beard=false), even though the labeling consistency experiment indicates that no beard could be assigned with >= 95% consistency. Selecting the mouth slightly open (MSO) for deeper analysis, we estimate the error rate for (MSO=true) at about 20% and (MSO=false) at about 2%. A corrected version of the MSO attribute values enables learning a model that achieves higher accuracy than previously reported for MSO. Corrected values for CelebA MSO are available at https://github.com/HaiyuWu/CelebAMSO.

Haiyu Wu, Sicong Tian, Huayu Li et al.

Ensuring logical consistency in predictions is a crucial yet overlooked aspect in multi-attribute classification. We explore the potential reasons for this oversight and introduce two pressing challenges to the field: 1) How can we ensure that a model, when trained with data checked for logical consistency, yields predictions that are logically consistent? 2) How can we achieve the same with data that hasn't undergone logical consistency checks? Minimizing manual effort is also essential for enhancing automation. To address these challenges, we introduce two datasets, FH41K and CelebA-logic, and propose LogicNet, an adversarial training framework that learns the logical relationships between attributes. Accuracy of LogicNet surpasses that of the next-best approach by 23.05%, 9.96%, and 1.71% on FH37K, FH41K, and CelebA-logic, respectively. In real-world case analysis, our approach can achieve a reduction of more than 50% in the average number of failed cases compared to other methods.

Aman Bhatta, Domingo Mery, Haiyu Wu et al.

State-of-the-art deep CNN face matchers are typically created using extensive training sets of color face images. Our study reveals that such matchers attain virtually identical accuracy when trained on either grayscale or color versions of the training set, even when the evaluation is done using color test images. Furthermore, we demonstrate that shallower models, lacking the capacity to model complex representations, rely more heavily on low-level features such as those associated with color. As a result, they display diminished accuracy when trained with grayscale images. We then consider possible causes for deeper CNN face matchers "not seeing color". Popular web-scraped face datasets actually have 30 to 60% of their identities with one or more grayscale images. We analyze whether this grayscale element in the training set impacts the accuracy achieved, and conclude that it does not. We demonstrate that using only grayscale images for both training and testing achieves accuracy comparable to that achieved using only color images for deeper models. This holds true for both real and synthetic training datasets. HSV color space, which separates chroma and luma information, does not improve the network's learning about color any more than in the RGB color space. We then show that the skin region of an individual's images in a web-scraped training set exhibits significant variation in their mapping to color space. This suggests that color carries limited identity-specific information. We also show that when the first convolution layer is restricted to a single filter, models learn a grayscale conversion filter and pass a grayscale version of the input color image to the next layer. Finally, we demonstrate that leveraging the lower per-image storage for grayscale to increase the number of images in the training set can improve accuracy of the face recognition model.

Patrick Tinsley, Sandip Purnapatra, Mahsa Mitcheff et al.

This paper describes the results of the 2023 edition of the ''LivDet'' series of iris presentation attack detection (PAD) competitions. New elements in this fifth competition include (1) GAN-generated iris images as a category of presentation attack instruments (PAI), and (2) an evaluation of human accuracy at detecting PAI as a reference benchmark. Clarkson University and the University of Notre Dame contributed image datasets for the competition, composed of samples representing seven different PAI categories, as well as baseline PAD algorithms. Fraunhofer IGD, Beijing University of Civil Engineering and Architecture, and Hochschule Darmstadt contributed results for a total of eight PAD algorithms to the competition. Accuracy results are analyzed by different PAI types, and compared to human accuracy. Overall, the Fraunhofer IGD algorithm, using an attention-based pixel-wise binary supervision network, showed the best-weighted accuracy results (average classification error rate of 37.31%), while the Beijing University of Civil Engineering and Architecture's algorithm won when equal weights for each PAI were given (average classification rate of 22.15%). These results suggest that iris PAD is still a challenging problem.

Sandipan Banerjee, Walter Scheirer, Kevin Bowyer et al.

In this article, we analyze how changing the underlying 3D shape of the base identity in face images can distort their overall appearance, especially from the perspective of deep face recognition. As done in popular training data augmentation schemes, we graphically render real and synthetic face images with randomly chosen or best-fitting 3D face models to generate novel views of the base identity. We compare deep features generated from these images to assess the perturbation these renderings introduce into the original identity. We perform this analysis at various degrees of facial yaw with the base identities varying in gender and ethnicity. Additionally, we investigate if adding some form of context and background pixels in these rendered images, when used as training data, further improves the downstream performance of a face recognition model. Our experiments demonstrate the significance of facial shape in accurate face matching and underpin the importance of contextual data for network training.

Haiyu Wu, Jaskirat Singh, Sicong Tian et al.

This paper studies how to synthesize face images of non-existent persons, to create a dataset that allows effective training of face recognition (FR) models. Besides generating realistic face images, two other important goals are: 1) the ability to generate a large number of distinct identities (inter-class separation), and 2) a proper variation in appearance of the images for each identity (intra-class variation). However, existing works 1) are typically limited in how many well-separated identities can be generated and 2) either neglect or use an external model for attribute augmentation. We propose Vec2Face, a holistic model that uses only a sampled vector as input and can flexibly generate and control the identity of face images and their attributes. Composed of a feature masked autoencoder and an image decoder, Vec2Face is supervised by face image reconstruction and can be conveniently used in inference. Using vectors with low similarity among themselves as inputs, Vec2Face generates well-separated identities. Randomly perturbing an input identity vector within a small range allows Vec2Face to generate faces of the same identity with proper variation in face attributes. It is also possible to generate images with designated attributes by adjusting vector values with a gradient descent method. Vec2Face has efficiently synthesized as many as 300K identities, whereas 60K is the largest number of identities created in the previous works. As for performance, FR models trained with the generated HSFace datasets, from 10k to 300k identities, achieve state-of-the-art accuracy, from 92% to 93.52%, on five real-world test sets (\emph{i.e.}, LFW, CFP-FP, AgeDB-30, CALFW, and CPLFW). For the first time, the FR model trained using our synthetic training set achieves higher accuracy than that trained using a same-scale training set of real face images on the CALFW, IJBB, and IJBC test sets.

Haiyu Wu, Sicong Tian, Aman Bhatta et al.

Reported face verification accuracy has reached a plateau on current well-known test sets. As a result, some difficult test sets have been assembled by reducing the image quality or adding artifacts to the image. However, we argue that test sets can be challenging without artificially reducing the image quality because the face recognition (FR) models suffer from correctly recognizing 1) the pairs from the same identity (i.e., genuine pairs) with a large face attribute difference, 2) the pairs from different identities (i.e., impostor pairs) with a small face attribute difference, and 3) the pairs of similar-looking identities (e.g., twins and relatives). We propose three challenging test sets to reveal important but ignored weaknesses of the existing FR algorithms. To challenge models on variation of facial attributes, we propose Hadrian and Eclipse to address facial hair differences and face exposure differences. The images in both test sets are high-quality and collected in a controlled environment. To challenge FR models on similar-looking persons, we propose twins-IND, which contains images from a dedicated twins dataset. The LFW test protocol is used to structure the proposed test sets. Moreover, we introduce additional rules to assemble "Goldilocks1" level test sets, including 1) restricted number of occurrence of hard samples, 2) equal chance evaluation across demographic groups, and 3) constrained identity overlap across validation folds. Quantitatively, without further processing the images, the proposed test sets have on-par or higher difficulties than the existing test sets. The datasets are available at: https: //github.com/HaiyuWu/SOTA-Face-Recognition-Train-and-Test.

Ying Qiu, Vítor Albiero, Michael C. King et al.

This paper is the first to explore the question of whether images that are classified incorrectly by a face analytics algorithm (e.g., gender classification) are any more or less likely to participate in an image pair that results in a face recognition error. We analyze results from three different gender classification algorithms (one open-source and two commercial), and two face recognition algorithms (one open-source and one commercial), on image sets representing four demographic groups (African-American female and male, Caucasian female and male). For impostor image pairs, our results show that pairs in which one image has a gender classification error have a better impostor distribution than pairs in which both images have correct gender classification, and so are less likely to generate a false match error. For genuine image pairs, our results show that individuals whose images have a mix of correct and incorrect gender classification have a worse genuine distribution (increased false non-match rate) compared to individuals whose images all have correct gender classification. Thus, compared to images that generate correct gender classification, images that generate gender classification errors do generate a different pattern of recognition errors, both better (false match) and worse (false non-match).

Priyanka Das, Joseph McGrath, Zhaoyuan Fang et al.

Launched in 2013, LivDet-Iris is an international competition series open to academia and industry with the aim to assess and report advances in iris Presentation Attack Detection (PAD). This paper presents results from the fourth competition of the series: LivDet-Iris 2020. This year's competition introduced several novel elements: (a) incorporated new types of attacks (samples displayed on a screen, cadaver eyes and prosthetic eyes), (b) initiated LivDet-Iris as an on-going effort, with a testing protocol available now to everyone via the Biometrics Evaluation and Testing (BEAT)(https://www.idiap.ch/software/beat/) open-source platform to facilitate reproducibility and benchmarking of new algorithms continuously, and (c) performance comparison of the submitted entries with three baseline methods (offered by the University of Notre Dame and Michigan State University), and three open-source iris PAD methods available in the public domain. The best performing entry to the competition reported a weighted average APCER of 59.10\% and a BPCER of 0.46\% over all five attack types. This paper serves as the latest evaluation of iris PAD on a large spectrum of presentation attack instruments.

Zhaoyuan Fang, Adam Czajka, Kevin W. Bowyer

Diversity and unpredictability of artifacts potentially presented to an iris sensor calls for presentation attack detection methods that are agnostic to specificity of presentation attack instruments. This paper proposes a method that combines two-dimensional and three-dimensional properties of the observed iris to address the problem of spoof detection in case when some properties of artifacts are unknown. The 2D (textural) iris features are extracted by a state-of-the-art method employing Binary Statistical Image Features (BSIF) and an ensemble of classifiers is used to deliver 2D modality-related decision. The 3D (shape) iris features are reconstructed by a photometric stereo method from only two images captured under near-infrared illumination placed at two different angles, as in many current commercial iris recognition sensors. The map of normal vectors is used to assess the convexity of the observed iris surface. The combination of these two approaches has been applied to detect whether a subject is wearing a textured contact lens to disguise their identity. Extensive experiments with NDCLD'15 dataset, and a newly collected NDIris3D dataset show that the proposed method is highly robust under various open-set testing scenarios, and that it outperforms all available open-source iris PAD methods tested in identical scenarios. The source code and the newly prepared benchmark are made available along with this paper.

Aidan Boyd, Adam Czajka, Kevin Bowyer

Gabor kernels are widely accepted as dominant filters for iris recognition. In this work we investigate, given the current interest in neural networks, if Gabor kernels are the only family of functions performing best in iris recognition, or if better filters can be learned directly from iris data. We use (on purpose) a single-layer convolutional neural network as it mimics an iris code-based algorithm. We learn two sets of data-driven kernels; one starting from randomly initialized weights and the other from open-source set of Gabor kernels. Through experimentation, we show that the network does not converge on Gabor kernels, instead converging on a mix of edge detectors, blob detectors and simple waves. In our experiments carried out with three subject-disjoint datasets we found that the performance of these learned kernels is comparable to the open-source Gabor kernels. These lead us to two conclusions: (a) a family of functions offering optimal performance in iris recognition is wider than Gabor kernels, and (b) we probably hit the maximum performance for an iris coding algorithm that uses a single convolutional layer, yet with multiple filters. Released with this work is a framework to learn data-driven kernels that can be easily transplanted into open-source iris recognition software (for instance, OSIRIS -- Open Source IRIS).

Vítor Albiero, Nisha Srinivas, Esteban Villalobos et al.

Matching live images (``selfies'') to images from ID documents is a problem that can arise in various applications. A challenging instance of the problem arises when the face image on the ID document is from early adolescence and the live image is from later adolescence. We explore this problem using a private dataset called Chilean Young Adult (CHIYA) dataset, where we match live face images taken at age 18-19 to face images on ID documents created at ages 9 to 18. State-of-the-art deep learning face matchers (e.g., ArcFace) have relatively poor accuracy for document-to-selfie face matching. To achieve higher accuracy, we fine-tune the best available open-source model with triplet loss for a few-shot learning. Experiments show that our approach achieves higher accuracy than the DocFace+ model recently developed for this problem. Our fine-tuned model was able to improve the true acceptance rate for the most difficult (largest age span) subset from 62.92% to 96.67% at a false acceptance rate of 0.01%. Our fine-tuned model is available for use by other researchers.

Joseph McGrath, Kevin W. Bowyer, Adam Czajka

This paper proposes the first, known to us, open source presentation attack detection (PAD) solution to distinguish between authentic iris images (possibly wearing clear contact lenses) and irises with textured contact lenses. This software can serve as a baseline in various PAD evaluations, and also as an open-source platform with an up-to-date reference method for iris PAD. The software is written in C++ and Python and uses only open source resources, such as OpenCV. This method does not incorporate iris image segmentation, which may be problematic for unknown fake samples. Instead, it makes a best guess to localize the rough position of the iris. The PAD-related features are extracted with the Binary Statistical Image Features (BSIF), which are classified by an ensemble of classifiers incorporating support vector machine, random forest and multilayer perceptron. The models attached to the current software have been trained with the NDCLD'15 database and evaluated on the independent datasets included in the LivDet-Iris 2017 competition. The software implements the functionality of retraining the classifiers with any database of authentic and attack images. The accuracy of the current version offered with this paper exceeds 99% when tested on subject-disjoint subsets of NDCLD'15, and oscillates around 85% when tested on the LivDet-Iris 2017 benchmarks, which is on par with the results obtained by the LivDet-Iris 2017 winner.

Siamul Karim Khan, Patrick Tinsley, Mahsa Mitcheff et al.

Synthesis of same-identity biometric iris images, both for existing and non-existing identities while preserving the identity across a wide range of pupil sizes, is complex due to the intricate iris muscle constriction mechanism, requiring a precise model of iris non-linear texture deformations to be embedded into the synthesis pipeline. This paper presents the first method of fully data-driven, identity-preserving, pupil size-varying synthesis of iris images. This approach is capable of synthesizing images of irises with different pupil sizes representing non-existing identities, as well as non-linearly deforming the texture of iris images of existing subjects given the segmentation mask of the target iris image. Iris recognition experiments suggest that the proposed deformation model both preserves the identity when changing the pupil size, and offers better similarity between same-identity iris samples with significant differences in pupil size, compared to state-of-the-art linear and non-linear (bio-mechanical-based) iris deformation models. Two immediate applications of the proposed approach are: (a) synthesis of, or enhancement of the existing biometric datasets for iris recognition, mimicking those acquired with iris sensors, and (b) helping forensic human experts examine iris image pairs with significant differences in pupil dilation. Images considered in this work conform to selected ISO/IEC 29794-6 quality metrics to make them applicable in biometric systems. The source codes and model weights are offered with this paper.

Gabriella Pangelinan, Grace Bezold, Haiyu Wu et al.

Facial brightness is a key image quality factor impacting face recognition accuracy differentials across demographic groups. In this work, we aim to decrease the accuracy gap between the similarity score distributions for Caucasian and African American female mated image pairs, as measured by d' between distributions. To balance brightness across demographic groups, we conduct three experiments, interpreting brightness in the face skin region either as median pixel value or as the distribution of pixel values. Balancing based on median brightness alone yields up to a 46.8% decrease in d', while balancing based on brightness distribution yields up to a 57.6% decrease. In all three cases, the similarity scores of the individual distributions improve, with mean scores maximally improving 5.9% for Caucasian females and 3.7% for African American females.

Sicong Tian, Haiyu Wu, Michael C. King et al.

One-to-many facial identification is documented to achieve high accuracy in the case where both the probe and the gallery are "mugshot quality" images. However, an increasing number of documented instances of wrongful arrest following one-to-many facial identification have raised questions about its accuracy. Probe images used in one-to-many facial identification are often cropped from frames of surveillance video and deviate from "mugshot quality" in various ways. This paper systematically explores how the accuracy of one-to-many facial identification is degraded by the person in the probe image choosing to wear dark sunglasses. We show that sunglasses degrade accuracy for mugshot-quality images by an amount similar to strong blur or noticeably lower resolution. Further, we demonstrate that the combination of sunglasses with blur or lower resolution results in even more pronounced loss in accuracy. These results have important implications for developing objective criteria to qualify a probe image for the level of accuracy to be expected if it used for one-to-many identification. To ameliorate the accuracy degradation caused by dark sunglasses, we show that it is possible to recover about 38% of the lost accuracy by synthetically adding sunglasses to all the gallery images, without model re-training. We also show that the frequency of wearing-sunglasses images is very low in existing training sets, and that increasing the representation of wearing-sunglasses images can greatly reduce the error rate. The image set assembled for this research is available at https://cvrl.nd.edu/projects/data/ to support replication and further research.

Haiyu Wu, Sicong Tian, Jacob Gutierrez et al.

A fundamental tenet of pattern recognition is that overlap between training and testing sets causes an optimistic accuracy estimate. Deep CNNs for face recognition are trained for N-way classification of the identities in the training set. Accuracy is commonly estimated as average 10-fold classification accuracy on image pairs from test sets such as LFW, CALFW, CPLFW, CFP-FP and AgeDB-30. Because train and test sets have been independently assembled, images and identities in any given test set may also be present in any given training set. In particular, our experiments reveal a surprising degree of identity and image overlap between the LFW family of test sets and the MS1MV2 training set. Our experiments also reveal identity label noise in MS1MV2. We compare accuracy achieved with same-size MS1MV2 subsets that are identity-disjoint and not identity-disjoint with LFW, to reveal the size of the optimistic bias. Using more challenging test sets from the LFW family, we find that the size of the optimistic bias is larger for more challenging test sets. Our results highlight the lack of and the need for identity-disjoint train and test methodology in face recognition research.

Xavier Merino, Gabriella Pangelinan, Samuel Langborgh et al.

Automated one-to-many (1:N) face recognition is a powerful investigative tool commonly used by law enforcement agencies. In this context, potential matches resulting from automated 1:N recognition are reviewed by human examiners prior to possible use as investigative leads. While automated 1:N recognition can achieve near-perfect accuracy under ideal imaging conditions, operational scenarios may necessitate the use of surveillance imagery, which is often degraded in various quality dimensions. One important quality dimension is image resolution, typically quantified by the number of pixels on the face. The common metric for this is inter-pupillary distance (IPD), which measures the number of pixels between the pupils. Low IPD is known to degrade the accuracy of automated face recognition. However, the threshold IPD for reliability in human face recognition remains undefined. This study aims to explore the boundaries of human recognition accuracy by systematically testing accuracy across a range of IPD values. We find that at low IPDs (10px, 5px), human accuracy is at or below chance levels (50.7%, 35.9%), even as confidence in decision-making remains relatively high (77%, 70.7%). Our findings indicate that, for low IPD images, human recognition ability could be a limiting factor to overall system accuracy.

Aman Bhatta, Maria Dhakal, Michael C. King et al.

A central problem in one-to-many facial identification is that the person in the probe image may or may not have enrolled image(s) in the gallery; that is, may be In-gallery or Out-of-gallery. Past approaches to detect when a rank-one result is Out-of-gallery have mostly focused on finding a suitable threshold on the similarity score. We take a new approach, using the additional enrolled images of the identity with the rank-one result to predict if the rank-one result is In-gallery / Out-of-gallery. Given a gallery of identities and images, we generate In-gallery and Out-of-gallery training data by extracting the ranks of additional enrolled images corresponding to the rank-one identity. We then train a classifier to utilize this feature vector to predict whether a rank-one result is In-gallery or Out-of-gallery. Using two different datasets and four different matchers, we present experimental results showing that our approach is viable for mugshot quality probe images, and also, importantly, for probes degraded by blur, reduced resolution, atmospheric turbulence and sunglasses. We also analyze results across demographic groups, and show that In-gallery / Out-of-gallery classification accuracy is similar across demographics. Our approach has the potential to provide an objective estimate of whether a one-to-many facial identification is Out-of-gallery, and thereby to reduce false positive identifications, wrongful arrests, and wasted investigative time. Interestingly, comparing the results of older deep CNN-based face matchers with newer ones suggests that the effectiveness of our Out-of-gallery detection approach emerges only with matchers trained using advanced margin-based loss functions.

Kagan Ozturk, Deeksha Arun, Kevin W. Bowyer et al.

Ear recognition has gained attention as a reliable biometric technique due to the distinctive characteristics of human ears. With the increasing availability of large-scale datasets, convolutional neural networks (CNNs) have been widely adopted to learn features directly from raw ear images, outperforming traditional hand-crafted methods. However, the effect of bilateral ear symmetry on the features learned by CNNs has received little attention in recent studies. In this paper, we investigate how bilateral ear symmetry influences the effectiveness of CNN-based ear recognition. To this end, we first develop an ear side classifier to automatically categorize ear images as either left or right. We then explore the impact of incorporating this side information during both training and test. Cross-dataset evaluations are conducted on five datasets. Our results suggest that treating left and right ears separately during training and testing can lead to notable performance improvements. Furthermore, our ablation studies on alignment strategies, input sizes, and various hyperparameter settings provide practical insights into training CNN-based ear recognition systems on large-scale datasets to achieve higher verification rates.

Kagan Ozturk, Louisa Conwill, Jacob Gutierrez et al.

Facial filters are now commonplace for social media users around the world. Previous work has demonstrated that facial filters can negatively impact automated face recognition performance. However, these studies focus on small numbers of hand-picked filters in particular styles. In order to more effectively incorporate the wide ranges of filters present on various social media applications, we introduce a framework that allows for larger-scale study of the impact of facial filters on automated recognition. This framework includes a controlled dataset of face images, a principled filter selection process that selects a representative range of filters for experimentation, and a set of experiments to evaluate the filters' impact on recognition. We demonstrate our framework with a case study of filters from the American applications Instagram and Snapchat and the Chinese applications Meitu and Pitu to uncover cross-cultural differences. Finally, we show how the filtering effect in a face embedding space can easily be detected and restored to improve face recognition performance.

Deeksha Arun, Kagan Ozturk, Kevin W. Bowyer et al.

Ear recognition has emerged as a promising biometric modality due to the relative stability in appearance during adulthood. Although Vision Transformers (ViTs) have been widely used in image recognition tasks, their efficiency in ear recognition has been hampered by a lack of attention to overlapping patches, which is crucial for capturing intricate ear features. In this study, we evaluate ViT-Tiny (ViT-T), ViT-Small (ViT-S), ViT-Base (ViT-B) and ViT-Large (ViT-L) configurations on a diverse set of datasets (OPIB, AWE, WPUT, and EarVN1.0), using an overlapping patch selection strategy. Results demonstrate the critical importance of overlapping patches, yielding superior performance in 44 of 48 experiments in a structured study. Moreover, upon comparing the results of the overlapping patches with the non-overlapping configurations, the increase is significant, reaching up to 10% for the EarVN1.0 dataset. In terms of model performance, the ViT-T model consistently outperformed the ViT-S, ViT-B, and ViT-L models on the AWE, WPUT, and EarVN1.0 datasets. The highest scores were achieved in a configuration with a patch size of 28x28 and a stride of 14 pixels. This patch-stride configuration represents 25% of the normalized image area (112x112 pixels) for the patch size and 12.5% of the row or column size for the stride. This study confirms that transformer architectures with overlapping patch selection can serve as an efficient and high-performing option for ear-based biometric recognition tasks in verification scenarios.

Aidan Boyd, Patrick Tinsley, Kevin W. Bowyer et al.

This work explores how human judgement about salient regions of an image can be introduced into deep convolutional neural network (DCNN) training. Traditionally, training of DCNNs is purely data-driven. This often results in learning features of the data that are only coincidentally correlated with class labels. Human saliency can guide network training using our proposed new component of the loss function that ConveYs Brain Oversight to Raise Generalization (CYBORG) and penalizes the model for using non-salient regions. This mechanism produces DCNNs achieving higher accuracy and generalization compared to using the same training data without human salience. Experimental results demonstrate that CYBORG applies across multiple network architectures and problem domains (detection of synthetic faces, iris presentation attacks and anomalies in chest X-rays), while requiring significantly less data than training without human saliency guidance. Visualizations show that CYBORG-trained models' saliency is more consistent across independent training runs than traditionally-trained models, and also in better agreement with humans. To lower the cost of collecting human annotations, we also explore using deep learning to provide automated annotations. CYBORG training of CNNs addresses important issues such as reducing the appetite for large training sets, increasing interpretability, and reducing fragility by generalizing better to new types of data.

Vítor Albiero, Kai Zhang, Michael C. King et al.

Media reports have accused face recognition of being ''biased'', ''sexist'' and ''racist''. There is consensus in the research literature that face recognition accuracy is lower for females, who often have both a higher false match rate and a higher false non-match rate. However, there is little published research aimed at identifying the cause of lower accuracy for females. For instance, the 2019 Face Recognition Vendor Test that documents lower female accuracy across a broad range of algorithms and datasets also lists ''Analyze cause and effect'' under the heading ''What we did not do''. We present the first experimental analysis to identify major causes of lower face recognition accuracy for females on datasets where previous research has observed this result. Controlling for equal amount of visible face in the test images mitigates the apparent higher false non-match rate for females. Additional analysis shows that makeup-balanced datasets further improves females to achieve lower false non-match rates. Finally, a clustering experiment suggests that images of two different females are inherently more similar than of two different males, potentially accounting for a difference in false match rates.

Andrey Kuehlkamp, Aidan Boyd, Adam Czajka et al.

Iris recognition of living individuals is a mature biometric modality that has been adopted globally from governmental ID programs, border crossing, voter registration and de-duplication, to unlocking mobile phones. On the other hand, the possibility of recognizing deceased subjects with their iris patterns has emerged recently. In this paper, we present an end-to-end deep learning-based method for postmortem iris segmentation and recognition with a special visualization technique intended to support forensic human examiners in their efforts. The proposed postmortem iris segmentation approach outperforms the state of the art and in addition to iris annulus, as in case of classical iris segmentation methods - detects abnormal regions caused by eye decomposition processes, such as furrows or irregular specular highlights present on the drying and wrinkling cornea. The method was trained and validated with data acquired from 171 cadavers, kept in mortuary conditions, and tested on subject-disjoint data acquired from 259 deceased subjects. To our knowledge, this is the largest corpus of data used in postmortem iris recognition research to date. The source code of the proposed method are offered with the paper. The test data will be available through the National Archive of Criminal Justice Data (NACJD) archives.

Aidan Boyd, Patrick Tinsley, Kevin Bowyer et al.

Can deep learning models achieve greater generalization if their training is guided by reference to human perceptual abilities? And how can we implement this in a practical manner? This paper proposes a training strategy to ConveY Brain Oversight to Raise Generalization (CYBORG). This new approach incorporates human-annotated saliency maps into a loss function that guides the model's learning to focus on image regions that humans deem salient for the task. The Class Activation Mapping (CAM) mechanism is used to probe the model's current saliency in each training batch, juxtapose this model saliency with human saliency, and penalize large differences. Results on the task of synthetic face detection, selected to illustrate the effectiveness of the approach, show that CYBORG leads to significant improvement in accuracy on unseen samples consisting of face images generated from six Generative Adversarial Networks across multiple classification network architectures. We also show that scaling to even seven times the training data, or using non-human-saliency auxiliary information, such as segmentation masks, and standard loss cannot beat the performance of CYBORG-trained models. As a side effect of this work, we observe that the addition of explicit region annotation to the task of synthetic face detection increased human classification accuracy. This work opens a new area of research on how to incorporate human visual saliency into loss functions in practice. All data, code and pre-trained models used in this work are offered with this paper.

Bingyu Shen, Brandon RichardWebster, Alice O'Toole et al.

Advances in face synthesis have raised alarms about the deceptive use of synthetic faces. Can synthetic identities be effectively used to fool human observers? In this paper, we introduce a study of the human perception of synthetic faces generated using different strategies including a state-of-the-art deep learning-based GAN model. This is the first rigorous study of the effectiveness of synthetic face generation techniques grounded in experimental techniques from psychology. We answer important questions such as how often do GAN-based and more traditional image processing-based techniques confuse human observers, and are there subtle cues within a synthetic face image that cause humans to perceive it as a fake without having to search for obvious clues? To answer these questions, we conducted a series of large-scale crowdsourced behavioral experiments with different sources of face imagery. Results show that humans are unable to distinguish synthetic faces from real faces under several different circumstances. This finding has serious implications for many different applications where face images are presented to human users.

Jeremy Speth, Nathan Vance, Patrick Flynn et al.

Remote photoplethysmography (rPPG) is a technique for estimating blood volume changes from reflected light without the need for a contact sensor. We present the first examples of presentation attacks in the digital and physical domains on rPPG from face video. Digital attacks are easily performed by adding imperceptible periodic noise to the input videos. Physical attacks are performed with illumination from visible spectrum LEDs placed in close proximity to the face, while still being difficult to perceive with the human eye. We also show that our attacks extend beyond medical applications, since the method can effectively generate a strong periodic pulse on 3D-printed face masks, which presents difficulties for pulse-based face presentation attack detection (PAD). The paper concludes with ideas for using this work to improve robustness of rPPG methods and pulse-based face PAD.

Jeremy Speth, Nathan Vance, Adam Czajka et al.

We present the Deception Detection and Physiological Monitoring (DDPM) dataset and initial baseline results on this dataset. Our application context is an interview scenario in which the interviewee attempts to deceive the interviewer on selected responses. The interviewee is recorded in RGB, near-infrared, and long-wave infrared, along with cardiac pulse, blood oxygenation, and audio. After collection, data were annotated for interviewer/interviewee, curated, ground-truthed, and organized into train / test parts for a set of canonical deception detection experiments. Baseline experiments found random accuracy for micro-expressions as an indicator of deception, but that saccades can give a statistically significant response. We also estimated subject heart rates from face videos (remotely) with a mean absolute error as low as 3.16 bpm. The database contains almost 13 hours of recordings of 70 subjects, and over 8 million visible-light, near-infrared, and thermal video frames, along with appropriate meta, audio and pulse oximeter data. To our knowledge, this is the only collection offering recordings of five modalities in an interview scenario that can be used in both deception detection and remote photoplethysmography research.

Aidan Boyd, Kevin Bowyer, Adam Czajka

Deep learning has driven remarkable accuracy increases in many computer vision problems. One ongoing challenge is how to achieve the greatest accuracy in cases where training data is limited. A second ongoing challenge is that trained models oftentimes do not generalize well even to new data that is subjectively similar to the training set. We address these challenges in a novel way, with the first-ever (to our knowledge) exploration of encoding human judgement about salient regions of images into the training data. We compare the accuracy and generalization of a state-of-the-art deep learning algorithm for a difficult problem in biometric presentation attack detection when trained on (a) original images with typical data augmentations, and (b) the same original images transformed to encode human judgement about salient image regions. The latter approach results in models that achieve higher accuracy and better generalization, decreasing the error of the LivDet-Iris 2020 winner from 29.78% to 16.37%, and achieving impressive generalization in a leave-one-attack-type-out evaluation scenario. This work opens a new area of study for how to embed human intelligence into training strategies for deep learning to achieve high accuracy and generalization in cases of limited training data.

KS Krishnapriya, Michael C. King, Kevin W. Bowyer

News reports have suggested that darker skin tone causes an increase in face recognition errors. The Fitzpatrick scale is widely used in dermatology to classify sensitivity to sun exposure and skin tone. In this paper, we analyze a set of manual Fitzpatrick skin type assignments and also employ the individual typology angle to automatically estimate the skin tone from face images. The set of manual skin tone rating experiments shows that there are inconsistencies between human raters that are difficult to eliminate. Efforts to automate skin tone rating suggest that it is particularly challenging on images collected without a calibration object in the scene. However, after the color-correction, the level of agreement between automated and manual approaches is found to be 96% or better for the MORPH images. To our knowledge, this is the first work to: (a) examine the consistency of manual skin tone ratings across observers, (b) document that there is substantial variation in the rating of the same image by different observers even when exemplar images are given for guidance and all images are color-corrected, and (c) compare manual versus automated skin tone ratings.

Shervin Minaee, Ping Luo, Zhe Lin et al.

Face detection is a crucial first step in many facial recognition and face analysis systems. Early approaches for face detection were mainly based on classifiers built on top of hand-crafted features extracted from local image regions, such as Haar Cascades and Histogram of Oriented Gradients. However, these approaches were not powerful enough to achieve a high accuracy on images of from uncontrolled environments. With the breakthrough work in image classification using deep neural networks in 2012, there has been a huge paradigm shift in face detection. Inspired by the rapid progress of deep learning in computer vision, many deep learning based frameworks have been proposed for face detection over the past few years, achieving significant improvements in accuracy. In this work, we provide a detailed overview of some of the most representative deep learning based face detection methods by grouping them into a few major categories, and present their core architectural designs and accuracies on popular benchmarks. We also describe some of the most popular face detection datasets. Finally, we discuss some current challenges in the field, and suggest potential future research directions.

Jeremy Speth, Nathan Vance, Patrick Flynn et al.

Remote photoplethysmography (rPPG), a family of techniques for monitoring blood volume changes, may be especially useful for widespread contactless health monitoring using face video from consumer-grade visible-light cameras. The COVID-19 pandemic has caused the widespread use of protective face masks. We found that occlusions from cloth face masks increased the mean absolute error of heart rate estimation by more than 80\% when deploying methods designed on unmasked faces. We show that augmenting unmasked face videos by adding patterned synthetic face masks forces the model to attend to the periocular and forehead regions, improving performance and closing the gap between masked and unmasked pulse estimation. To our knowledge, this paper is the first to analyse the impact of face masks on the accuracy of pulse estimation and offers several novel contributions: (a) 3D CNN-based method designed for remote photoplethysmography in a presence of face masks, (b) two publicly available pulse estimation datasets acquired from 86 unmasked and 61 masked subjects, (c) evaluations of handcrafted algorithms and a 3D CNN trained on videos of unmasked faces and with masks synthetically added, and (d) data augmentation method to add a synthetic mask to a face video.

Vítor Albiero, Kevin W. Bowyer

Recent news articles have accused face recognition of being "biased", "sexist" or "racist". There is consensus in the research literature that face recognition accuracy is lower for females, who often have both a higher false match rate and a higher false non-match rate. However, there is little published research aimed at identifying the cause of lower accuracy for females. For instance, the 2019 Face Recognition Vendor Test that documents lower female accuracy across a broad range of algorithms and datasets also lists "Analyze cause and effect" under the heading "What we did not do". We present the first experimental analysis to identify major causes of lower face recognition accuracy for females on datasets where previous research has observed this result. Controlling for equal amount of visible face in the test images reverses the apparent higher false non-match rate for females. Also, principal component analysis indicates that images of two different females are inherently more similar than of two different males, potentially accounting for a difference in false match rates.

Aidan Boyd, Zhaoyuan Fang, Adam Czajka et al.

As the popularity of iris recognition systems increases, the importance of effective security measures against presentation attacks becomes paramount. This work presents an overview of the most important advances in the area of iris presentation attack detection published in recent two years. Newly-released, publicly-available datasets for development and evaluation of iris presentation attack detection are discussed. Recent literature can be seen to be broken into three categories: traditional "hand-crafted" feature extraction and classification, deep learning-based solutions, and hybrid approaches fusing both methodologies. Conclusions of modern approaches underscore the difficulty of this task. Finally, commentary on possible directions for future research is provided.

Kevin W. Bowyer, Michael King, Walter Scheirer et al.

The automatic analysis of face images can generate predictions about a person's gender, age, race, facial expression, body mass index, and various other indices and conditions. A few recent publications have claimed success in analyzing an image of a person's face in order to predict the person's status as Criminal / Non-Criminal. Predicting criminality from face may initially seem similar to other facial analytics, but we argue that attempts to create a criminality-from-face algorithm are necessarily doomed to fail, that apparently promising experimental results in recent publications are an illusion resulting from inadequate experimental design, and that there is potentially a large social cost to belief in the criminality from face illusion.

Kai Zhang, Vítor Albiero, Kevin W. Bowyer

Web-scraped, in-the-wild datasets have become the norm in face recognition research. The numbers of subjects and images acquired in web-scraped datasets are usually very large, with number of images on the millions scale. A variety of issues occur when collecting a dataset in-the-wild, including images with the wrong identity label, duplicate images, duplicate subjects and variation in quality. With the number of images being in the millions, a manual cleaning procedure is not feasible. But fully automated methods used to date result in a less-than-ideal level of clean dataset. We propose a semi-automated method, where the goal is to have a clean dataset for testing face recognition methods, with similar quality across men and women, to support comparison of accuracy across gender. Our approach removes near-duplicate images, merges duplicate subjects, corrects mislabeled images, and removes images outside a defined range of pose and quality. We conduct the curation on the Asian Face Dataset (AFD) and VGGFace2 test dataset. The experiments show that a state-of-the-art method achieves a much higher accuracy on the datasets after they are curated. Finally, we release our cleaned versions of both datasets to the research community.