Kevin W. Bowyer - Face Recognition

Lessons from Collecting a Million Biometric Samples,
P. Jonathon Phillips, Patrick J. Flynn and Kevin W. Bowyer,
Image and Vision Computing Journal 58, 96–107, February 2017.
DOI link to this paper in ScienceDirect.
Over the past decade, independent evaluations have become commonplace in many areas of experimental computer science, including face and gesture recognition. A key attribute of many successful independent evaluations is a curated data set. Desired aspects associated with these data sets include appropriateness to the experimental design, a corpus size large enough to allow statistically rigorous characterization of results, and the availability of comprehensive metadata that allow inferences to be made on various data set attributes. In this paper, we review a ten-year biometric sampling effort that enabled the creation of several key biometrics challenge problems. We summarize the design and execution of data collections, identify key challenges, and convey some lessons learned.

SREFI: Synthesis of Realistic Example Face Images,
Sandipan Banerjee, John S. Bernhard, Walter J. Scheirer, Kevin W. Bowyer and Patrick J. Flynn,
IEEE International Joint Conference on Biometrics, October 2017, Denver.
DOI link to this paper in IEEE Xplore.
In this paper, we propose a novel face synthesis approach that can generate an arbitrarily large number of synthetic images of both real and synthetic identities. Thus a face image dataset can be expanded in terms of the number of identities represented and the number of images per identity using this approach, without the identity-labeling and privacy complications that come from downloading images from the web. To measure the visual fidelity and uniqueness of the synthetic face images and identities, we conducted face matching experiments with both human participants and a CNN pre-trained on a dataset of 2.6M real face images. To evaluate the stability of these synthetic faces, we trained a CNN model with an augmented dataset containing close to 200,000 synthetic faces. We used a snapshot of this trained CNN to recognize extremely challenging frontal (real) face images. Experiments showed training with the augmented faces boosted the face recognition performance of the CNN.

Demography-based Facial Retouching Detection using Subclass Supervised Sparse Autoencoder,
Aparna Bharati, Mayank Vatsa, Richa Singh and Kevin W. Bowyer, Xin Tong,
IEEE International Joint Conference on Biometrics, October 2017, Denver.
DOI link to this paper in IEEE Xplore.
Digital retouching of face images is becoming more widespread due to the introduction of software packages that automate the task. Several researchers have introduced algorithms to detect whether a face image is original or retouched. However, previous work on this topic has not considered whether or how accuracy of retouching detection varies with the demography of face images. In this paper, we introduce a new Multi-Demographic Retouched Faces (MDRF) dataset, which contains images belonging to two genders, male and female, and three ethnicities, Indian, Chinese, and Caucasian. Further, retouched images are created using two different retouching software packages. The second major contribution of this research is a novel semi-supervised autoencoder incorporating "subclass" information to improve classification. The proposed approach outperforms existing state-of-the-art detection algorithms for the task of generalized retouching detection. Experiments conducted with multiple combinations of ethnicities show that accuracy of retouching detection can vary greatly based on the demographics of the training and testing images.

Biometric Identification of Identical Twins: A Survey
Kevin W. Bowyer and Patrick J. Flynn,
IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS), September 2016.
pdf of this paper.
The ability of biometric techniques to distinguish between identical twins is of interest for multiple reasons. The research literature touching on this topic is spread across a variety of areas. This survey pulls together the literature to date in this area, identifies available datasets for research, points out topics of uncertainty and suggests possible future research.

Template Aging in 3D and 2D Face Recognition,
Ishan Manjani, Hakki Sumerkan, Patrick J. Flynn and Kevin W. Bowyer,
IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS), September 2016.
pdf of this paper.
This is the first work to explore template aging in 3D face recognition. We use a dataset of images representing 16 subjects with 3D and 2D face images, and compare shortterm and long-term time-lapse matching accuracy. We find that an ensemble-of-regions approach to 3D face matching has much greater accuracy than whole-face 3D matching, or than a commercial 2D matcher. We observe a drop in accuracies with increased time lapse, most with whole-face 3D matching followed by 2D matching and the 3D ensemble of regions approach. Finally, we determine whether the difference in match quality arising with an increased time lapse is statistically significant.

On accuracy estimation and comparison of results in biometric research,
Domingo Mery, Yuning Zhao and Kevin Bowyer,
IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS), September 2016.
pdf of this paper.
The estimated accuracy of an algorithm is the most important element of the typical biometrics research publication. Comparisons between algorithms are commonly made based on estimated accuracies reported in different publications. However, even when the same dataset is used in two publications, there is a very low frequency of the publications using the same protocol for estimating algorithm accuracy. Using the example problems of face recognition, expression recognition and gender classification, we show that the variation in estimated performance on the same dataset across different protocols can be enormous. Based on these results, we make recommendations for how to obtain performance estimates that allow reliable comparison between algorithms.

Detecting Facial Retouching Using Supervised Deep Learning,
Aparna Bharati, Richa Singh,Mayank Vatsa and Kevin W. Bowyer,
IEEE Transactions on Information Forensics and Security, September 2016, 1903-1913.
pdf of this paper.
Digitally altering, or "retouching", face images is a common practice for images on social media, photo sharing websites, and even identification cards when the standards are not strictly enforced. This research demonstrates the effect of digital alterations on the performance of automatic face recognition, and also introduces an algorithm to classify face images as original or retouched with good high performance. We first introduce two face image databases with unaltered and retouched images. Face recognition experiments performed on these databases show that when a retouched image is matched with its original image or an unaltered gallery image, the identification performance is considerably degraded, with a drop in matching accuracy of up to 25%. However, when images are retouched with the same style, the matching accuracy can be misleadingly high in comparison to matching original images. To detect retouching in face images, a novel supervised deep Boltzmann machine algorithm is proposed. It uses facial parts to learn discriminative features to classify face images as original or retouched. The proposed approach for classifying images as original or retouched yields an accuracy of over 87% on the datasets introduced in this paper and over 99% on three other makeup datasets used by previous researchers. This is a substantial increase in accuracy over the previous stateof- the-art algorithm [5] which has shown less than 50% accuracy in classifying original and retouched images from the ND-IIITD Retouched Faces database

Automatic facial attribute analysis via adaptive sparse representation of random patches,
Domingo Mery and Kevin W. Bowyer,
Pattern Recognition Letters 68 (2), 260-269, December 2015.
pdf of this paper.
... This paper addresses the problem of automated recognition of facial attributes by proposing a new general approach called Adaptive Sparse Representation of Random Patches (ASR+). ... Experiments were carried out on eight face databases in order to recognize facial expression, gender, race, disguise and beard. Results show that ASR+ deals well with unconstrained conditions, outperforming various representative methods in the literature in many complex scenarios.

Near-IR to Visible Light Face Matching: Effectiveness of Pre-Processing Options for Commercial Matchers,
John Bernhard, Jeremiah Barr, Kevin W. Bowyer and Patrick J. Flynn
IEEE International Conference on Biometrics Theory, Applications and Systems (BTAS 2015), to appear.
pdf of this paper.
... Image pre-processing techniques can potentially be used to help reduce the differences between near-IR and visible light images, with the goal of improving matching accuracy. We evaluate the use of several such techniques in combination with commercial matchers and show that simply extracting the red plane results in a comparable improvement in accuracy. In addition, we show that many of the pre-processing techniques hinder the ability of existing commercial matchers to extract templates. We also make available a new dataset called Near Infrared Visible Light Database (ND-NIVL) consisting of visible light and near- IR face images with accompanying baseline performance for several commercial matchers.

Face Recognition under Pose Variation with Local Gabor Features Enhanced by Active Shape and Statistical Models,
Leonardo A. Cament, Francisco J. Galdames, Kevin W. Bowyer, and Claudio Perez,
Pattern Recognition 48 (11), 3371-3384, November 2015.
pdf of this paper.

Location Matters: A Study of the Effects of Environment on Facial Recognition for Biometric Security,
Amanda Sgroi, Patrick Flynn, Kevin W. Bowyer and Hannah Garvey,
Biometrics in the Wild Workshop 2015 (BWild 2015), Ljubljana, Slovenia, May 2015.
pdf of this paper.

Lessons from collecting a million biometric samples,
P. Jonathon Phillips, Patrick J. Flynn and Kevin W. Bowyer,
IEEE International Conference on Automatic Face and Gesture Recognition (FG 2015), Ljubljana, Slovenia, May 2015.
pdf of this paper.
Over the past decade, independent evaluations have become commonplace in many areas of experimental computer science, including face and gesture recognition. A key attribute of many successful independent evaluations is a curated data set. Desired things associated with these data sets include appropriateness to the experimental design, a corpus size large enough to allow statistically rigorous characterization of results, and the availability of comprehensive metadata that allow inferences to be made on various data set attributes. In this paper, we review a ten-year biometric sampling effort that enabled the creation of several key biometrics challenge problems. We summarize the design and execution of data collections, identify key challenges, and convey some lessons learned.

Strong, Neutral or Weak: Exploring the Impostor Score Distribution,
Amanda Sgroi, Patrick Flynn, Kevin Bowyer and P. Jonathon Phillips,
IEEE Transactions on Information Forensics and Security, 10 (6), 1207-1220, June 2015.
pdf of this paper.
The strong, neutral, or weak (SNoW) face impostor pairs problem is intended to explore the causes and impact of impostor face pairs that are inherently strong (easily recognized as nonmatches) or weak (possible false matches). The SNoW technique develops three partitions within the impostor score distribution of a given data set. Results provide evidence that varying degrees of impostor scores impact the overall performance of a face recognition system. This paper extends our earlier work to incorporate improvements regarding outlier detection for partitioning, explores the SNoW concept for the additional modalities of fingerprint and iris, and presents methods for how to begin to reveal the causes of weak impostor pairs. We also show a clear operational difference between strong and weak comparisons as well as identify partition stability across multiple algorithms.

Framework for Active Clustering with Ensembles,
Jeremiah R. Barr, Kevin W. Bowyer and Patrick J. Flynn,
IEEE Transactions on Information Forensics and Security, 9 (11) 1986-2001, November 2014.
pdf of this paper.
... We introduce a novel semi-supervised framework for clustering face patterns into identity groups using minimal human interaction. This technique combines concepts from ensemble clustering and active learning to improve clustering accuracy. The framework actively queries the user for a soft link constraint between each pair of neighboring faces that are ambiguously matched according to the ensemble. We demonstrate the efficacy of our approach with the broadest evaluation of active face clustering algorithms to date. ...

Recognition of Facial Attributes Using Adaptive Sparse Representations of Random Patches,
Domingo Mery and Kevin Bowyer,
ECCV Workshop on Soft Biometrics, September 2014.
pdf of this paper.
... This paper addresses the problem of automated recognition of facial attributes by proposing a new general approach called Adap- tive Sparse Representation of Random Patches (ASR+). ...
Best Paper Award.

Active Clustering with Ensembles for Social Structure Extraction,
Jeremiah Barr, Leonardo Cament, Kevin Bowyer and Patrick Flynn,
IEEE Winter Conference on Applications of Computer Vision (WACV), March 2014.
pdf of this paper.
We introduce a method for extracting the social network structure for the persons appearing in a set of video clips. Individuals are unknown, and are not matched against known enrollments. An identity cluster representing an individual is formed by grouping similar-appearing faces from different videos. Each identity cluster is represented by a node in the social network. Two nodes are linked if the faces from their clusters appeared together in one or more video frames. Our approach incorporates a novel active clustering technique to create more accurate identity clusters based on feedback from the user about ambiguously matched faces. The final output consists of one or more network structures that represent the social group(s), and a list of persons who potentially connect multiple social groups. Our results demonstrate the efficacy of the proposed clustering algorithm and network analysis techniques.

Jeffrey R. Paone, Patrick J. Flynn, P. Jonathon Philips, Kevin W. Bowyer, Richard W. Vorder Bruegge, Patrick J. Grother, George W. Quinn, Matthew T. Pruitt and Jason M. Grant,
Double Trouble: Differentiating Identical Twins by Face Recognition,
IEEE Transactions in Information Forensics and Security 9 (2), 285-295, February 2014.
link to this paper in IEEE Xplore.
Facial recognition algorithms should be able to operate even when similar-looking individuals are encountered, or even in the extreme case of identical twins. An experimental data set comprised of 17486 images from 126 pairs of identical twins (252 subjects) collected on the same day and 6864 images from 120 pairs of identical twins (240 subjects) with images taken a year later was used to measure the performance on seven different face recognition algorithms. Performance is reported for variations in illumination, expression, gender, and age for both the same day and cross-year image sets. Regardless of the conditions of image acquisition, distinguishing identical twins are significantly harder than distinguishing subjects who are not identical twins for all algorithms.

The Challenge of Face Recognition from Digital Point-and-Shoot Cameras,
Ross Beveridge, Jonathon Phillips, David Bolme, Bruce Draper, Geof Givens, Yui Man Lui, Mohammad Nayeem Teli, Hao Zhang, W. Todd Scruggs, Kevin Bowyer, Patrick Flynn and Su Cheng,
Biometrics Theory, Applications and Systems (BTAS), Sept 30 - Oct 2, 2013.
pdf of this paper.
Inexpensive point-and-shoot camera technology has combined with social network technology to give the general population a motivation to use face recognition technology. ... this paper introduces the Point-and-Shoot Face Recognition Challenge (PaSC). The challenge includes 9,376 still images of 293 people balanced with respect to distance to the camera, alternative sensors, frontal versus not-frontal views, and varying location. There are also 2,802 videos for 265 people: a subset of the 293. Verification results are presented for public baseline algorithms and a commercial algorithm for three cases: comparing still images to still images, videos to videos, and still images to videos.

SNoW: Understanding the Causes of Strong, Neutral and Weak Face Impostor Pairs,
Amanda Sgroi, Kevin W. Bowyer, Patrick Flynn and P Jonathon Phillips,
Biometrics Theory, Applications and Systems (BTAS), Sept 30 - Oct 2, 2013.
pdf of this paper.
The Strong, Neutral, or Weak Face Impostor Pairs problem was generated to explore the causes and impact of impostor face pairs that span varying strengths of scores. We develop three partitions within the impostor distribution for a given algorithm. The Strong partition contains image pairs that are easy to categorize as impostors. The Neutral partition contains image pairs that are less easily categorized as impostors. The Weak partition contains image pairs that are likely to cause false positives. ...

Pose-Robust Recognition of Low-Resolution Face Images,
Soma Biswas, Gaurav Aggarwal, Patrick J. Flynn and Kevin W. Bowyer,
IEEE Transactions on Pattern Analysis and Machine Intelligence, 35 (12), 3037-3049, December 2013.
pdf of this paper.
Face images captured by surveillance cameras usually have poor resolution in addition to uncontrolled poses and illumination conditions, all of which adversely affect performance of face matching algorithms. In this paper, we develop a completely automatic, novel approach for matching surveillance quality facial images to high resolution images in frontal pose which are often available during enrollment. The proposed approach uses multidimensional scaling to simultaneously transform the features from the poor quality probe images and the high quality gallery images in such a manner that the distances between them approximate the distances had the probe images been captured in the same conditions as the gallery images. Tensor analysis is used for facial landmark localization in the low-resolution uncontrolled probe images for computing the features. Thorough evaluation on the Multi-PIE dataset and comparisons with state-of-the-art super-resolution and classifierbased approaches are performed to illustrate the usefulness of the proposed approach. Experiments on surveillance imagery further signify the applicability of the framework. We also show the usefulness of the proposed approach for the application of tracking and recognition in surveillance videos.

Face Recognition from Video: A Review,
Jeremiah Barr, Kevin W. Bowyer, Patrick Flynn, Soma Biswas,
International Journal of Pattern Recognition and Artificial Intelligence 26 (5), August 2012.
pdf of this paper.
... present a broad and deep review of recently proposed methods for overcoming the difficulties encountered in unconstrained settings. We also draw connections between the ways in which humans and current algorithms recognize faces. An overview of the most popular and difficult publicly available face video databases is provided to complement these discussions. Finally, we cover key research challenges and opportunities that lie ahead for the field as a whole.

Fusion of Face and Iris Biometrics,
Ryan Connaughton, Kevin W. Bowyer and Patrick Flynn,
in Handbook of Iris Recognition, Mark Burge and Kevin W. Bowyer, editors, Springer, 2012.
pdf of this chapter.

Human and Machine Performance on Periocular Biometrics Under Near-Infrared Light and Visible Light,
Karen P. Hollingsworth, Shelby S. Darnell, Philip E. Miller, Damon L. Woodard, Kevin W. Bowyer and Patrick J. Flynn,
IEEE Transactions on Information Forensics and Security 7 (2), 588-601, April 2012.
pdf of this paper.
... Previous periocular research has used either visible light or near-infrared light images, but no prior research has directly compared the two illuminations using images with similar resolution. We conducted an experiment in which volunteers were asked to compare pairs of periocular images. ... We calculated performance of three computer algorithms on the periocular images. Performance for humans and computers was similar.

Best of Face and Gesture Recognition 2011,
Rainer Stiefelhagen, Marian Stewart Bartlett and Kevin W. Bowyer,
Image and Vision Computing Journal 30 (3), 135, March 2012.
Link to this overview of the special issue.

A Sparse Representation Approach to Face Matching Across Plastic Surgery,
Gaurav Aggarwal, Soma Biswas, Patrick J. Flynn and Kevin W. Bowyer,
IEEE Workshop on Applications of Computer Vision, January 2012, Colorado Springs, CO.
pdf of this paper.
... In this paper, we propose a novel approach to address the challenges involved in automatic matching of faces across plastic surgery variations. .... Extensive experiments conducted on a recently introduced plastic surgery database consisting of 900 subjects highlight the effectiveness of the proposed approach.

Predicting Good, Bad and Ugly Match Pairs,
Gaurav Aggarwal, Soma Biswas, Patrick J. Flynn and Kevin W. Bowyer,
IEEE Workshop on Applications of Computer Vision, January 2012, Colorado Springs, CO.
pdf of this paper.
... The recently introduced GBU challenge problems indicates that even when the impact of most known factors is eliminated or significantly reduced by the data collection and experimental protocol, there can be significant variation in performance across different partitions of the data. ... In this paper, we investigate various image and facial characteristics that can account for the observed and significant difference in performance across these partitions. ...

Useful Features for Human Verification in Near-Infrared Periocular Images,
Karen Hollingsworth, Kevin W. Bowyer and Patrick J. Flynn,
Image and Vision Computing Journal 29 (11), October 2011, 707-715.
pdf of this paper.
... We conducted two experiments to determine how humans analyze periocular images. In these experiments, we presented pairs of images and asked volunteers to determine whether the two images showed eyes from the same subject or from different subjects. ...
Reprinted in the Journal of Intelligence Community Research and Development.

Multidimensional Scaling for Matching Low-resolution Face Images,
Soma Biswas, Kevin W. Bowyer and Patrick J. Flynn,
IEEE Transactions on Pattern Analysis and Machine Intelligence 34 (10),October 2012, 2019-2030..
pdf of this paper.
... we propose a novel approach for matching low resolution probe images with higher resolution gallery images ... The proposed method simultaneously embeds the low resolution probe images and the high resolution gallery images in a common space such that the distances between them in the transformed space approximates the distances had both the images been of high resolution. The two mappings are learned simultaneously from high resolution training images using iterative majorization algorithm. Extensive evaluation of the proposed approach on the Multi-PIE dataset with probe image resolution as low as 8 x 6 pixels illustrates the usefulness of the method. We show that the proposed approach improves the matching performance significantly as compared to performing matching in the low-resolution domain or using super-resolution techniques to obtain a higher-resolution test image prior to recognition. ...

A Study of Face Recognition of Identical Twins By Humans,
Soma Biswas, Kevin W. Bowyer and Patrick J. Flynn,
International Workshop on Information Forensics and Security (WIFS 2011), December 2011 Foz do Iguacu, Brazil.
pdf of this paper.
In this work, we investigate human capability to distinguish between identical twins. If humans are able to distinguish between facial images of identical twins, it would suggest that humans are capable of identifying discriminating facial traits that can potentially be useful to develop algorithms for this very challenging problem. Experiments with different viewing times and imaging conditions are conducted to determine if humans viewing a pair of facial images can perceive if the image pairs belong to the same person or to a pair of identical twins. The experiments are conducted on 186 twin subjects, making it the largest such study in the literature to date.

Twins 3D Face Recognition Challenge,
Vipin Vijayan, Kevin W. Bowyer, Patrick Flynn, Di Huang, Liming Chen, Mark Hansen, Omar Ocegueda, Shishir Shah, Ioannis Kakadiaris,
International Joint Conference on Biometrics (IJCB 2011), October 2011.
pdf of this paper.
Existing 3D face recognition algorithms have achieved high enough performances against public datasets like FRGC v2, that it is difficult to achieve further significant increases in recognition performance. However, the 3D TEC dataset is a more challenging dataset which consists of 3D scans of 107 pairs of twins that were acquired in a single session, with each subject having a scan of a neutral expression and a smiling expression. The combination of factors related to the facial similarity of identical twins and the variation in facial expression makes this a challenging dataset. We conduct experiments using state of the art face recognition algorithms and present the results. Our results indicate that 3D face recognition of identical twins in the presence of varying facial expressions is far from a solved problem, but that good performance is possible.

What Surprises Do Identical Twins Have for Identity Science?,
Kevin W. Bowyer,
IEEE Computer 44 (7), July 2011, 100-102.
DOI link to this paper.
Experiments with biometric datasets from identical twins are helping to shape future research in face and iris recognition.

Distinguishing Identical Twins By Face Recognition,
P. Jonathon Phillips, Patrick J. Flynn, Kevin W. Bowyer, Richard W. Vorder Bruegge, Patrick J. Grother, George W. Quinn, Matthew Pruitt,
IEEE International Conference on Automatic Face and Gesture Recognition (FG 2011), March 2011, 185-192.
pdf of this paper.
This paper measures ability of face recognition algorithms to distinguish between identical twin siblings. The experimental dataset consists of images taken of 126 pairs of identical twins (252 people) collected on the same day and 24 pairs of identical twins (48 people) with images collected one year apart. Recognition experiments are conducted using three of the top submissions to the Multiple Biometric Evaluation (MBE) 2010 Still Face Track. ...

Detecting Questionable Observers Using Face Track Clustering,
Jeremiah Barr, Kevin W. Bowyer and Patrick J. Flynn,
IEEE Computer Society Workshop on Applications of Computer Vision, January 2011.
pdf of this paper.
We introduce the questionable observer detection problem: Given a collection of videos of crowds, determine which individuals appear unusually often across the set of videos. The algorithm proposed here detects these individuals by clustering sequences of face images. ...

Identifying Useful Features for Recognition In Near-infrared Periocular Images,
Karen Hollingsworth, Kevin W. Bowyer and Patrick J Flynn,
Fourth IEEE International Conference on Biometrics Theory, Applications and Systems (BTAS 10), 2010.
pdf of this paper.
... We presented pairs of periocular images to testers and asked them to determine whether the two images were from the same person or from different people. Our testers correctly determined the relationship in over 90% of the queries. We asked them to describe what features in the images were helpful to them in making their decisions. We found that eyelashes, tear ducts, shape of the eye, and eyelids were used most frequently in determining whether two images were of the same person. ...

Introduction to the Special Issue on Recent Advances In Biometrics,
Kevin W. Bowyer,
IEEE Transactions on Systems, Man and Cybernetics - Part A, 40 (3), May 2010, 434-436.
pdf of this paper.
As with BTAS 07, a biometrics-themed special issue of Systems Man and Cybernetics - Part A was organized following BTAS 08. However, unlike the SMC-A special section drawn from BTAS 07, this special issue had an "open" call for papers, meaning that submissions were not limited to papers presented at BTAS 08. ... A total of 31 submissions were received for the special issue. ... The ten papers that appear in this special issue represent the result of this process. Four of the ten papers appearing in this special issue are revised and extended versions of papers presented in the closing session of the BTAS 08 conference, a session which, by BTAS tradition, is reserved for the submissions that receive the overall best reviews from the conference program committee. The papers in this special issue cover a range of different biometric modalities, including face, ear, iris, signature and multi-modal. In the area of face recognition, there are papers dealing with 2D, 3D, and hand-drawn sketches. Also, the papers in this special issue range from relatively application oriented to relatively theoretical. One common theme is that each paper addresses an important current topic in biometrics research and makes a novel contribution to the state of the art.

FRVT 2006 and ICE 2006 Large-Scale Experimental Results,
P. Jonathon Phillips, W. Todd Scruggs, Alice O'Toole, Patrick J. Flynn, Kevin W. Bowyer, Cathy L. Schott and Matthew Sharpe,
IEEE Transactions on Pattern Analysis and Machine Intelligence 32 (5), May 2010, 831-846.
pdf of this paper.
DOI link.
This paper describes the large-scale experimental results from the Face Recognition Vendor Test (FRVT) 2006 and the Iris Challenge Evaluation (ICE) 2006. ...

Overview of the Multiple Biometric Grand Challenge,
P. Jonathon Phillips, Patrick J. Flynn, J. Ross Beveridge, W. Todd Scruggs, Alice J. O'Toole, David Bolme, Kevin W. Bowyer, Bruce A. Draper, Geof. H. Givens, Yui Man Lui, Hassan Sahibzada, Joseph A. Scallan, and Samuel Weimer.
International Conference on Biometrics, June 2009, 705-714.
pdf of this paper.
The goal of the Multiple Biometric Grand Challenge (MBGC) is to improve the performance of face and iris recognition technology from samples acquired under unconstrained conditions. The MBGC is organized into three challenge problems. Each challenge problem relaxes the constraints in different directions. ...

Introduction to the Special Section of Best Papers from the 2007 Biometrics: Theory, Applications and Systems Conference,
Kevin W. Bowyer,
IEEE Transactions on Systems, Man and Cybernetics - Part A, 39 (1), January 2009, 2-3.
pdf of this paper.
DOI link.
... Over 100 papers were submitted to BTAS 07. ... The final result of this process is the set of five papers that appear in this special section. We are particularly fortunate in the way that the five papers in this special section illustrate the breadth of activities in current biometrics research. Face, fingerprint, iris, voice, handwriting, and multimodal biometrics are all represented. ...

Using Multi-Instance Enrollment to Improve Performance of 3D Face Recognition,
Timothy C. Faltemier, Kevin W. Bowyer and Patrick J. Flynn,
Computer Vision and Image Understanding 112 (2), November 2008, 114-125.
Preprint pdf version of this paper.
DOI link.
This paper explores the use of multi-instance enrollment as a means to improve the performance of 3D face recognition. Experiments are performed using the ND-2006 3D face data set which contains 13,450 scans of 888 subjects. This is the largest 3D face data set currently available and contains a substantial amount of varied facial expression. Results indicate that the multi-instance enrollment outperforms a state-of-the-art component-based recognition approach ...

Multi-factor Approach To Improving Recognition Performance In Surveillance-quality Video,
Deborah Thomas, Kevin W. Bowyer and Patrick J. Flynn,
Biometrics: Theory, Applications and Systems (BTAS 08), September 2008, Washington, DC.
DOI link.

Profile Face Detection: A Subset Multi-biometric Approach,
James Gentile, Kevin W. Bowyer and Patrick J. Flynn,
Biometrics: Theory, Applications and Systems (BTAS 08), September 2008, Washington, DC.
DOI link.

A Region Ensemble for 3D Face Recognition,
Timothy Faltemier, Kevin W. Bowyer and Patrick J. Flynn,
IEEE Transactions on Information Forensics and Security, 3(1):62-73, March 2008.
DOI link.
... we introduce a new system for 3D face recognition based on the fusion of results from a committee of regions that have been independently matched. ... Rank-one recognition rates of 97.2% and verification rates of 93.2% at 0.1% false accept rate are reported and compared to other methods published on the Face Recognition Grand Challenge v2 data set.

Guest Editorial: Introduction to the Special Issue on Recent Advances in Biometric Systems,
Kevin W. Bowyer, Venu Govindaraju and Nalini Ratha,
IEEE Transactions on Systems, Man and Cybernetics - B 37 (5), October 2007.
pdf of this paper.
We are pleased to present 14 papers in this special issue devoted to recent advances in biometric systems. ...

FRVT 2006 and ICE 2006 Large-Scale Results,
P. J. Phillips, W. T. Scruggs, A. J. O'Toole, P. J. Flynn, K.W. Bowyer, C. L. Schott, and M. Sharpe.
National Institute of Standards and Technology, NISTIR 7408, http://face.nist.gov, 2007.
pdf of this report.
This report describes the large-scale experimental results from the Face Recognition Vendor Test (FRVT) 2006 and the Iris Challenge Evaluation (ICE) 2006. ...
Our research group is part of the Multiple Biometric Grand Challenge and Iris Challenge Evaluation support teams.

A Fast Algorithm for ICP-based 3D Shape Biometrics,
Ping Yan and Kevin W. Bowyer,
Computer Vision and Image Understanding, 107 (3), 195-202, September 2007.
pdf of this paper.
... we present a novel approach, called "Pre-computed Voxel Nearest Neighbor," to reduce the computational time for shape matching in a biometrics context. The approach shifts the heavy computation burden to the enrollment stage, which is done offline. Experiments in 3D ear biometrics with 369 subjects and 3D face biometrics with 219 subjects demonstrate the effectiveness of our approach.

Actively Exploring Face Space(s) for Improved Face Recognition,
Nitesh V. Chawla and Kevin W. Bowyer,
AAAI 2007, Vancouver, July 2007.
pdf of this paper.
We propose a learning framework that actively explores creation of face space(s) by selecting images that are complementary to the images already represented in the face space. We also construct ensembles of classifiers learned from such actively sampled image sets, which further provides improvement in the recognition rates. ...

Face Recognition Using 2D, 3D and Infra-Red: Is Multi-modal Better than Multi-sample?
Kevin W. Bowyer, Kyong I. Chang, Patrick J. Flynn and Xin Chen.
Proceedings of the IEEE, 94 (11), 2000-2012, November 2006.
pdf of this paper.
We compare the performance improvement obtained by combining three-dimensional or infra-red with normal intensity images (a multi-modal approach) to the performance improvement obtained by using multiple intensity images (a multi-sample approach). Combining results from different types of imagery gives significantly higher recognition rates than are obtained by using a single intensity image. However, significantly higher recognition rates are also obtained by combining results from multiple intensity images.

Multiple Nose Region Matching for 3D Face Recognition Under Varying Facial Expression,
Kyong I. Chang, Kevin W. Bowyer, and Patrick J. Flynn,
IEEE Transactions on Pattern Analysis and Machine Intelligence, 28 (10), 1695-1700, October 2006.
pdf of this paper.
An algorithm is proposed for 3D face recognition in the presence of varied facial expressions. It is based on combining the match scores from matching multiple overlapping regions around the nose. Experimental results are presented using the largest database employed to date in 3D face recognition studies, over 4,000 scans of 449 subjects. ...

Multi-modal Biometrics: An Overview,
Kevin W. Bowyer, et al,
Second Workshop on Multi-Modal User Authentication (MMUA 2006), May 2006, Toulouse, France.
pdf of this paper.
The topic of multi-modal biometrics has attracted strong interest in recent years. This paper categorizes approaches to multi-modal biometrics based on the biometric source, the type of sensing used, and the depth of collaborative interaction in the processing. This paper also attempts to identify some of the challenges and issues that confront research in multi-modal biometrics.
This paper represents the invited talk given to open the first day of the workshop.

A Survey of Approaches and Challenges in 3D and Multi-modal 3D+2D Face Recognition,
Kevin W. Bowyer, Kyong Chang, and Patrick J. Flynn,
Computer Vision and Image Understanding 101 (1), January 2006, 1-15.
pdf of this paper.
... This survey focuses on face recognition performed by matching models of the three-dimensional shape of the face, either alone or in combination with matching corresponding two-dimensional intensity images.
This article was number one on the CVIU "Top 25" list for the quarters October - December 2005 and January - March of 2006, and in the top ten for seven consecutive quarters.

Eye Perturbation Approach for Robust Recognition of Inaccurately Aligned Faces,
Jaesik Min, Kevin W. Bowyer, and Patrick J. Flynn,
Fifth International Conference on Audio- and Video-Based Biometric Person Authentication (AVBPA 2005), July 2005, New York, 41-50.
pdf of this paper.
... For improved performance and robustness to the eye location variation, we propose an eye perturbation approach that generates multiple face extractions from a query image by using the perturbed eye locations centered at the initial eye locations. ...

Infra-Red and Visible-Light Face Recognition,
Xin Chen, Patrick J. Flynn, and Kevin W. Bowyer,
Computer Vision and Image Understanding 99 (3), September 2005, 332-358.
pdf of this paper.
... We find that in a scenario involving time lapse between gallery and probe, and relatively controlled lighting, (1) PCA-based recognition using visible images outperforms PCA-based recognition using infra-red images, and (2) the combination of PCA-based recognition using visible and infra-red imagery substantially outperforms either one individually...
This article was number two on the CVIU "Top 25" list for July - September of 2005.

Empirical Evaluation of Advanced Ear Biometrics,
Ping Yan and Kevin W. Bowyer,
Workshop on Empirical Evaluation Methods in Computer Vision, San Diego, CA, June 2005.
pdf of this paper.
We present results of the largest experimental investigation of ear biometrics to date. Approaches considered include a PCA ("eigen-ear") approach with 2D intensity images, achieving 63.8% rank-one recognition; a PCA approach with range images, achieving 55.3% Hausdorff matching of edge images from range images, achieving 67.5% and ICP matching of the 3D data, achieving 98.7%. ICP based matching not only achieves the best performance, but also shows good scalability with size of dataset. The data set used represents over 300 persons, each with images acquired on at least two different dates. In addition, the ICP-based approach is further applied on an expanded data set of 404 subjects, and achieves 97.5% rank one recognition rate. In order to test the robustness and variability of ear biometrics, ear symmetry is also investigated. In our experiments around 90% of peoples' right and left ears are symmetric.

Overview of the Face Recognition Grand Challenge,
P. Jonathon Phillips, Patrick J. Flynn, Todd Scruggs, Kevin W. Bowyer, Jin Chang, Kevin Hoffman, Joe Marques, Jaesik Min, and William Worek,
Computer Vision and Pattern Recognition (CVPR 2005), San Diego, June 2005, I:947-954.
pdf of this paper.
Our research group is part of the Face Recognition Grand Challenge support team.

Random Subspaces and Subsampling for 2-D Face Recognition,
Nitesh V. Chawla and Kevin W. Bowyer,
Computer Vision and Pattern Recognition (CVPR 2005) , San Diego, June 2005, II: 582-589.
pdf of this paper.
Random subspaces are a popular ensemble construction technique that improves the accuracy of weak classifiers. It has been shown, in different domains, that random subspaces combined with weak classifiers such as decision trees and nearest neighbor classifiers can provide an improvement in accuracy. In this paper, we apply the random subspace methodology to the 2-D face recognition task. The main goal of the paper is to see if the random subspace methodology can do as well, if not better, than the single classifier constructed on the tuned face space. We also propose the use of a validation set for tuning the face space, to avoid bias in the accuracy estimation. In addition, we also compare the random subspace methodology to an ensemble of subsamples of image data. This work shows that a random subspaces ensemble can outperform a well-tuned single classifier for a typical 2-D face recognition problem. The random subspaces approach has the added advantage of requiring less careful tweaking.

An Evaluation of Multi-modal 2D+3D Face Biometrics,
Kyong I. Chang, Kevin W. Bowyer, and Patrick J. Flynn,
IEEE Transactions on Pattern Analysis and Machine Intelligence 27 (4), April 2005, 619-624.
pdf of this paper.
We report on the largest experimental study to date in multi-modal 2D+3D face recognition ... Major conclusions are: (1) 2D and 3D have similar recognition performance when considered individually, (2) Combining 2D and 3D results using a simple weighting scheme outperforms either 2D or 3D alone, (3) Combining results from two or more 2D images using a similar weighting scheme also outperforms a single 2D image, and (4) Combined 2D+3D outperforms the multi-image 2D result. This is the first (so far, only) work to present such an experimental control to substantiate multi-modal performance improvement."

Face Recognition Technology and the Security Versus Privacy Tradeoff,
Kevin W. Bowyer,
IEEE Technology and Society, Spring 2004, 9-20.
pdf of this paper.
Video surveillance and face recognition systems have become the subject of increased interest and controversy after the September 11 terrorist attacks on the United States. ... This paper analyzes the interplay of technical and social issues involved in the widespread application of video surveillance for person identification.
This paper received a 2005 Award of Excellence from the Society for Technical Communication.

Assessment of Time Dependency in Face Recognition: An Initial Study,
Patrick Flynn, Kevin W. Bowyer and P. Jonathon Phillips,
Audio- and Video-Based Biometric Person Authentication (AVBPA 2003), Springer Lecture Notes in Computer Science 2688, 44-51.
pdf of this paper.
... Experimental results suggest that (a) recognition performance is substantially poorer when unknown images are acquired on a different day from the enrolled images, (b) degradation in performance does not follow a simple predictable pattern with time between known and unknown image acquisition, and (c) performance figures quoted in the literature based on known and unknown image sets acquired on the same day may have little practical value.

Comparison and Combination of Ear and Face Images for Appearance-based Biometrics,
Kyong Chang, Kevin W. Bowyer, Sudeep Sarkar, and Barnabas Victor,
IEEE Transactions on Pattern Analysis and Machine Intelligence 25 (9), September 2003, 1160-1165.
pdf of this paper.
In the experiments reported here, recognition performance is essentially identical using ear images or face images and combining the two for multimodal recognition results in a statistically significant performance improvement. ... To our knowledge, ours is the only work to present any experimental results of computer algorithms for biometric recognition based on the ear.

Face Recognition Using 2D and 3D Facial Data,
Kyong I. Chang, Kevin W. Bowyer and Patrick J. Flynn,
First Workshop on Multi-Modal User Authentication , Santa Barbara, 25-32, December 2003.
pdf of this paper.
Results are presented for the largest experimental study to date that investigates the comparison and combination of 2D and 3D face recognition. To our knowledge, this is the only such study to incorporate significant time lapse between gallery and probe image acquisition ...