Learning Boltzmann distance metric for face recognition

We introduce a new method for face recognition using a versatile probabilistic model known as Restricted Boltzmann Machine (RBM). In particular, we propose to regularise the standard data likelihood learning with an information-theoretic distance metric defined on intra-personal images. This results in an effective face representation which captures the regularities in the face space and minimises the intra-personal variations. In addition, our method allows easy incorporation of multiple feature sets with controllable level of sparsity. Our experiments on a high variation dataset show that the proposed method is competitive against other metric learning rivals. We also investigated the RBM method under a variety of settings, including fusing facial parts and utilising localised feature detectors under varying resolutions. In particular, the accuracy is boosted from 71.8% with the standard whole-face pixels to 99.2% with combination of facial parts, localised feature extractors and appropriate resolutions.

Making the most of the self-quotient image in face recognition

The self-quotient image is a biologically inspired representation which has been proposed as an illumination invariant feature for automatic face recognition. Owing to the lack of strong domain specific assumptions underlying this representation, it can be readily extracted from raw images irrespective of the persons's pose, facial expression etc. What makes the self-quotient image additionally attractive is that it can be computed quickly and in a closed form using simple low-level image operations. However, it is generally accepted that the self-quotient is insufficiently robust to large illumination changes which is why it is mainly used in applications in which low precision is an acceptable compromise for high recall (e.g. retrieval systems). Yet, in this paper we demonstrate that the performance of this representation in challenging illuminations has been greatly underestimated. We show that its error rate can be reduced by over an order of magnitude, without any changes to the representation itself. Rather, we focus on the manner in which the dissimilarity between two self-quotient images is computed. By modelling the dominant sources of noise affecting the representation, we propose and evaluate a series of different dissimilarity measures, the best of which reduces the initial error rate of 63.0% down to only 5.7% on the notoriously challenging YaleB data set.

Illumination-invariant face recognition from a single image across extreme pose using a dual dimension AAM ensemble in the thermal infrared spectrum

Over the course of the last decade, infrared (IR) and particularly thermal IR imaging based face recognition has emerged as a promising complement to conventional, visible spectrum based approaches which continue to struggle when applied in practice. While inherently insensitive to visible spectrum illumination changes, IR data introduces specific challenges of its own, most notably sensitivity to factors which affect facial heat emission patterns, e.g. emotional state, ambient temperature, and alcohol intake. In addition, facial expression and pose changes are more difficult to correct in IR images because they are less rich in high frequency detail which is an important cue for fitting any deformable model. In this paper we describe a novel method which addresses these major challenges. Specifically, when comparing two thermal IR images of faces, we mutually normalize their poses and facial expressions by using an active appearance model (AAM) to generate synthetic images of the two faces with a neutral facial expression and in the same view (the average of the two input views). This is achieved by piecewise affine warping which follows AAM fitting. A major contribution of our work is the use of an AAM ensemble in which each AAM is specialized to a particular range of poses and a particular region of the thermal IR face space. Combined with the contributions from our previous work which addressed the problem of reliable AAM fitting in the thermal IR spectrum, and the development of a person-specific representation robust to transient changes in the pattern of facial temperature emissions, the proposed ensemble framework accurately matches faces across the full range of yaw from frontal to profile, even in the presence of scale variation (e.g. due to the varying distance of a subject from the camera). The effectiveness of the proposed approach is demonstrated on the largest public database of thermal IR images of faces and a newly acquired data set of thermal IR motion videos. Our approach achieved perfect recognition performance on both data sets, significantly outperforming the current state of the art methods even when they are trained with multiple images spanning a range of head views.

Infrared face recognition: a literature review

Automatic face recognition (AFR) is an area with immense practical potential which includes a wide range of commercial and law enforcement applications, and it continues to be one of the most active research areas of computer vision. Even after over three decades of intense research, the state-of-the-art in AFR continues to improve, benefiting from advances in a range of different fields including image processing, pattern recognition, computer graphics and physiology. However, systems based on visible spectrum images continue to face challenges in the presence of illumination, pose and expression changes, as well as facial disguises, all of which can significantly decrease their accuracy. Amongst various approaches which have been proposed in an attempt to overcome these limitations, the use of infrared (IR) imaging has emerged as a particularly promising research direction. This paper presents a comprehensive and timely review of the literature on this subject.

Vesselness features and the inverse compositional AAM for robust face recognition using thermal IR

Over the course of the last decade, infrared (IR) and particularly thermal IR imaging based face recognition has emerged as a promising complement to conventional, visible spectrum based approaches which continue to struggle when applied in the real world. While inherently insensitive to visible spectrum illumination changes, IR images introduce specific challenges of their own, most notably sensitivity to factors which affect facial heat emission patterns, e.g. emotional state, ambient temperature, and alcohol intake. In addition, facial expression and pose changes are more difficult to correct in IR images because they are less rich in high frequency detail which is an important cue for fitting any deformable model. In this paper we describe a novel method which addresses these major challenges. Specifically, to normalize for pose and facial expression changes we generate a synthetic frontal image of a face in a canonical, neutral facial expression from an image of the face in an arbitrary pose and facial expression. This is achieved by piecewise affine warping which follows active appearance model (AAM) fitting. This is the first publication which explores the use of an AAM on thermal IR images; we propose a pre-processing step which enhances detail in thermal images, making AAM convergence faster and more accurate. To overcome the problem of thermal IR image sensitivity to the exact pattern of facial temperature emissions we describe a representation based on reliable anatomical features. In contrast to previous approaches, our representation is not binary; rather, our method accounts for the reliability of the extracted features. This makes the proposed representation much more robust both to pose and scale changes. The effectiveness of the proposed approach is demonstrated on the largest public database of thermal IR images of faces on which it achieved 100% identification rate, significantly outperforming previously described methods

Towards person authentication by fusing visual and thermal face biometrics

In this chapter we focus on face appearance-based biometrics. The cheap and readily available hardware used to acquire data, their non-invasiveness and the ease of employing them from a distance and without the awareness of the user, are just some of the reasons why these continue to be of great practical interest. However, a number of research challenges remain. Specifically, face biometrics have traditionally focused on images acquired in the visible light spectrum and these are greatly affected by such extrinsic factors such as the illumination, camera angle (or, equivalently, head pose) and occlusion. In practice, the effects of changing pose are usually least problematic and can oftentimes be overcome by acquiring data over a time period, e.g., by tracking a face in a surveillance video. Consequently, image sequence or image set matching has recently gained a lot of attention in the literature [137–139] and is the paradigm adopted in this chapter as well. In other words, we assume that the training image set for each individual contains some variability in pose, but is not obtained in scripted conditions or in controlled illumination. In contrast, illumination is much more difficult to deal with: the illumination setup is in most cases not practical to control and its physics is difficult to accurately model. Thermal spectrum imagery is useful in this regard as it is virtually insensitive to illumination changes, as illustrated in Fig. 6.1. On the other hand, it lacks much of the individual, discriminating facial detail contained in visual images. In this sense, the two modalities can be seen as complementing each other. The key idea behind the system presented in this chapter is that robustness to extreme illumination changes can be achieved by fusing the two. This paradigm will further prove useful when we consider the difficulty of recognition in the presence of occlusion caused by prescription glasses.

Unfolding a face : from singular to manifold

Face recognition from a single image remains an important task in many practical applications and a significant research challenge. Some of the challenges are inherent to the problem, for example due to changing lighting conditions. Others, no less significant, are of a practical nature – face recognition algorithms cannot be assumed to operate on perfect data, but rather often on data that has already been subject to pre-processing errors (e.g. localization and registration errors). This paper introduces a novel method for face recognition that is both trained and queried using only a single image per subject. The key concept, motivated by abundant prior work on face appearance manifolds, is that of face part manifolds – it is shown that the appearance seen through a sliding window overlaid over an image of a face, traces a trajectory over a 2D manifold embedded in the image space. We present a theoretical argument for the use of this representation and demonstrate how it can be effectively exploited in the single image based recognition. It is shown that while inheriting the advantages of local feature methods, it also implicitly captures the geometric relationship between discriminative facial features and is naturally robust to face localization errors. Our theoretical arguments are verified in an experimental evaluation on the Yale Face Database.

Gradient edge map features for frontal face recognition under extreme illumination changes

Our aim in this paper is to robustly match frontal faces in the presence of extreme illumination changes, using only a single training image per person and a single probe image. In the illumination conditions we consider, which include those with the dominant light source placed behind and to the side of the user, directly above and pointing downwards or indeed below and pointing upwards, this is a most challenging problem. The presence of sharp cast shadows, large poorly illuminated regions of the face, quantum and quantization noise and other nuisance effects, makes it difficult to extract a sufficiently discriminative yet robust representation. We introduce a representation which is based on image gradient directions near robust edges which correspond to characteristic facial features. Robust edges are extracted using a cascade of processing steps, each of which seeks to harness further discriminative information or normalize for a particular source of extra-personal appearance variability. The proposed representation was evaluated on the extremely difficult YaleB data set. Unlike most of the previous work we include all available illuminations, perform training using a single image per person and match these also to a single probe image. In this challenging evaluation setup, the proposed gradient edge map achieved 0.8% error rate, demonstrating a nearly perfect receiver-operator characteristic curve behaviour. This is by far the best performance achieved in this setup reported in the literature, the best performing methods previously proposed attaining error rates of approximately 6–7%.

Multi-sensory face biometric fusion (for personal identification)

The objective of this work is to recognize faces using sets of images in visual and thermal spectra. This is challenging because the former is greatly affected by illumination changes, while the latter frequently contains occlusions due to eye-wear and is inherently less discriminative. Our method is based on a fusion of the two modalities. Specifically: we examine (i) the effects of preprocessing of data in each domain, (ii) the fusion of holistic and local facial appearance, and (iii) propose an algorithm for combining the similarity scores in visual and thermal spectra in the presence of prescription glasses and significant pose variations, using a small number of training images (5-7). Our system achieved a high correct identification rate of 97% on a freely available test set of 29 individuals and extreme illumination changes.

Image reduction operators based on non-monotonic averaging functions

Image reduction is a crucial task in image processing, underpinning many practical applications. This work proposes novel image reduction operators based on non-monotonic averaging aggregation functions. The technique of penalty function minimisation is used to derive a novel mode-like estimator capable of identifying the most appropriate pixel value for representing a subset of the original image. Performance of this aggregation function and several traditional robust estimators of location are objectively assessed by applying image reduction within a facial recognition task. The FERET evaluation protocol is applied to confirm that these non-monotonic functions are able to sustain task performance compared to recognition using nonreduced images, as well as significantly improve performance on query images corrupted by noise. These results extend the state of the art in image reduction based on aggregation functions and provide a basis for efficiency and accuracy improvements in practical computer vision applications.

Anthropogenic selection enhances cancer evolution in Tasmanian devil tumours

The Tasmanian Devil Facial Tumour Disease (DFTD) provides a unique opportunity to elucidate the long-term effects of natural and anthropogenic selection on cancer evolution. Since first observed in 1996, this transmissible cancer has caused local population declines by >90%. So far, four chromosomal DFTD variants (strains) have been described and karyotypic analyses of 253 tumours showed higher levels of tetraploidy in the oldest strain. We propose that increased ploidy in the oldest strain may have evolved in response to effects of genomic decay observed in asexually reproducing organisms. In this study, we focus on the evolutionary response of DFTD to a disease suppression trial. Tumours collected from devils subjected to the removal programme showed accelerated temporal evolution of tetraploidy compared with tumours from other populations where no increase in tetraploid tumours were observed. As ploidy significantly reduces tumour growth rate, we suggest that the disease suppression trial resulted in selection favouring slower growing tumours mediated by an increased level of tetraploidy. Our study reveals that DFTD has the capacity to rapidly respond to novel selective regimes and that disease eradication may result in novel tumour adaptations, which may further imperil the long-term survival of the world's largest carnivorous marsupial.

Identification of natural killer cell receptor genes in the genome of the marsupial Tasmanian devil (Sarcophilus harrisii)

Within the mammalian immune system, natural killer (NK) cells contribute to the first line of defence against infectious agents and tumours. Their activity is regulated, in part, by cell surface NK cell receptors. NK receptors can be divided into two unrelated, but functionally analogous superfamilies based on the structure of their extracellular ligand-binding domains. Receptors belonging to the C-type lectin superfamily are predominantly encoded in the natural killer complex (NKC), while receptors belonging to the immunoglobulin superfamily are predominantly encoded in the leukocyte receptor complex (LRC). Natural killer cell receptors are emerging as a rapidly evolving gene family which can display significant intra- and interspecific variation. To date, most studies have focused on eutherian mammals, with significantly less known about the evolution of these receptors in marsupials. Here, we describe the identification of 43 immunoglobulin domain-containing LRC genes in the genome of the Tasmanian devil (Sarcophilus harrisii), the largest remaining marsupial carnivore and only the second marsupial species to be studied. We also identify orthologs of NKC genesKLRK1, CD69, CLEC4E, CLEC1B, CLEC1A and an ortholog of an opossum NKC receptor. Characterisation of these regions in a second, distantly related marsupial provides new insights into the dynamic evolutionary histories of these receptors in mammals. Understanding the functional role of these genes is also important for the development of therapeutic agents against Devil Facial Tumour Disease, a contagious cancer that threatens the Tasmanian devil with extinction.

Major histocompatibility complex (MHC) markers in conservation biology

Human impacts through habitat destruction, introduction of invasive species and climate change are increasing the number of species threatened with extinction. Decreases in population size simultaneously lead to reductions in genetic diversity, ultimately reducing the ability of populations to adapt to a changing environment. In this way, loss of genetic polymorphism is linked with extinction risk. Recent advances in sequencing technologies mean that obtaining measures of genetic diversity at functionally important genes is within reach for conservation programs. A key region of the genome that should be targeted for population genetic studies is the Major Histocompatibility Complex (MHC). MHC genes, found in all jawed vertebrates, are the most polymorphic genes in vertebrate genomes. They play key roles in immune function via immune-recognition and -surveillance and host-parasite interaction. Therefore, measuring levels of polymorphism at these genes can provide indirect measures of the immunological fitness of populations. The MHC has also been linked with mate-choice and pregnancy outcomes and has application for improving mating success in captive breeding programs. The recent discovery that genetic diversity at MHC genes may protect against the spread of contagious cancers provides an added impetus for managing and protecting MHC diversity in wild populations. Here we review the field and focus on the successful applications of MHC-typing for conservation management. We emphasize the importance of using MHC markers when planning and executing wildlife rescue and conservation programs but stress that this should not be done to the detriment of genome-wide diversity.

Telomere dynamics and homeostasis in a transmissible cancer

Devil Facial Tumour Disease (DFTD) is a unique clonal cancer that threatens the world's largest carnivorous marsupial, the Tasmanian devil (Sarcophilus harrisii) with extinction. This transmissible cancer is passed between individual devils by cell implantation during social interactions. The tumour arose in a Schwann cell of a single devil over 15 years ago and since then has expanded clonally, without showing signs of replicative senescence; in stark contrast to a somatic cell that displays a finite capacity for replication, known as the “Hayflick limit”.