Nicholas Costen

Spatial Content and Spatial Quantisation Effects in Face Recognition

It has recently become apparent that if face images are degraded by spatial quantisation, or block averaging, there is a nonlinear acceleration of the decline in accuracy of recognition as block size increases. This suggests recognition requires a critical minimum range of object spatial frequencies. Two experiments were performed to clarify the phenomenon. In experiment 1, the speed and accuracy of recognition for six frontoparallel photographs of faces were measured. After familiarisation training sessions, the images were shown for 100 ms with 11, 21, and 42 pixels per face, horizontally measured. Transformations calculated to remove the same range of spatial frequencies were performed by means of quantisation, a Fourier low-pass filter, and Gaussian blurring. Although accuracy declined and speed increased in a significant, nonlinear manner in all cases as the image quality was reduced, it did so at a faster rate for the quantised images. In experiment 2, faces rated as being typical were shown at 9, 12, 23, and 45 pixels per face and with appropriate Fourier low-pass versions. The nonlinear decline was confirmed and it was shown that it could not be attributed to a ceiling effect. A further condition allowed quantised and Fourier low-pass conditions to be compared with an unstructured-noise condition of equal strength to that of the quantised images. These gave comparable, but slightly less impaired, recognition than the quantised images. It can be inferred from these results that the removal of a critical range of at least 8–16 cycles per face of information explains the step decline in recognition seen with quantised images. However, the decline found with quantised images is reinforced by internal masking from pixelisation.

How should we represent faces for automatic recognition

We describe results obtained from a testbed used to investigate different codings for automatic face recognition. An eigenface coding of shape-free faces using manually located landmarks was more effective than the corresponding coding of correctly shaped faces. Configuration also proved an effective method of recognition, with rankings given to incorrect matches relatively uncorrelated with those from shape-free faces. Both sets of information combine to improve significantly the performance of either system. The addition of a system, which directly correlated the intensity values of shape-free images, also significantly increased recognition, suggesting extra information was still available. The recognition advantage for shape-free faces reflected and depended upon high-quality representation of the natural facial variation via a disjoint ensemble of shape-free faces; if the ensemble comprised nonfaces, a shape-free disadvantage was induced. Manipulation within the shape-free coding to emphasize distinctive features of the faces, by caricaturing, allowed further increases in performance; this effect was only noticeable when the independent shape-free and configuration coding was used. Taken together, these results strongly support the suggestion that faces should be considered as lying in a high-dimensional manifold, which is locally linearly approximated by these shapes and textures, possibly with a separate system for local features. Principal components analysis is then seen as a convenient tool in this local approximation.

An Evaluation of Video-to-Video Face Verification

Person recognition using facial features, e.g., mug-shot images, has long been used in identity documents. However, due to the widespread use of web-cams and mobile devices embedded with a camera, it is now possible to realize facial video recognition, rather than resorting to just still images. In fact, facial video recognition offers many advantages over still image recognition; these include the potential of boosting the system accuracy and deterring spoof attacks. This paper presents an evaluation of person identity verification using facial video data, organized in conjunction with the International Conference on Biometrics (ICB 2009). It involves 18 systems submitted by seven academic institutes. These systems provide for a diverse set of assumptions, including feature representation and preprocessing variations, allowing us to assess the effect of adverse conditions, usage of quality information, query selection, and template construction for video-to-video face authentication.

Automatic Face Recognition: What Representation?

A testbed for automatic face recognition shows an eigenface coding of shape-free texture, with manually coded landmarks, was more effective than correctly shaped faces, being dependent upon high-quality representation of the facial variation by a shape-free ensemble. Configuration also allowed recognition, these measures combine to improve performance and allowed automatic measurement of the face-shape. Caricaturing further increased performance. Correlation of contours of shapefree images also increased recognition, suggesting extra information was available. A natural model considers faces as in a manifold, linearly approximated by the two factors, with a separate system for local features.

Sparse models for gender classification

A class of sparse regularization functions is considered for the developing sparse classifiers for determining facial gender. The sparse classification method aims to both select the most important features and maximize the classification margin, in a manner similar to support vector machines. An efficient process for directly calculating the complete set of optimal, sparse classifiers is developed. A single classification hyper-plane, which maximizes posterior probability of describing training data, is then efficiently selected. The classifier is tested on a Japanese gender-divided ensemble, described via a collection of appearance models. Performance is comparable with a linear SVM, and allows effective manipulation of apparent gender.

Automated regional analysis of B-mode ultrasound images of skeletal muscle movement.

To understand the functional significance of skeletal muscle anatomy, a method of quantifying local shape changes in different tissue structures during dynamic tasks is required. Taking advantage of the good spatial and temporal resolution of B-mode ultrasound imaging, we describe a method of automatically segmenting images into fascicle and aponeurosis regions and tracking movement of features, independently, in localized portions of each tissue. Ultrasound images (25 Hz) of the medial gastrocnemius muscle were collected from eight participants during ankle joint rotation (2° and 20°), isometric contractions (1, 5, and 50 Nm), and deep knee bends. A Kanade-Lucas-Tomasi feature tracker was used to identify and track any distinctive and persistent features within the image sequences. A velocity field representation of local movement was then found and subdivided between fascicle and aponeurosis regions using segmentations from a multiresolution active shape model (ASM). Movement in each region was quantified by interpolating the effect of the fields on a set of probes. ASM segmentation results were compared with hand-labeled data, while aponeurosis and fascicle movement were compared with results from a previously documented cross-correlation approach. ASM provided good image segmentations (<1 mm average error), with fully automatic initialization possible in sequences from seven participants. Feature tracking provided similar length change results to the cross-correlation approach for small movements, while outperforming it in larger movements. The proposed method provides the potential to distinguish between active and passive changes in muscle shape and model strain distributions during different movements/conditions and quantify nonhomogeneous strain along aponeuroses.

Masking of faces by facial and non-facial stimuli

Abstract A visual masking technique was used as a tool to study the recognition of well-known faces. Famous faces, but unknown masks were used. The visual relation between target and mask was varied; the effect was measured in terms of the thresholds for reports of “familiarity” and the ability to name the target. Three experiments are reported, all of which employed a backward-masking paradigm. The target and mask had equal display times, and there was no interstimulus interval. Significant masking was found with face masks but none from either object or noise masks. Intermediate levels of masking were found from faces that were inverted, jumbled, or had their inner features removed. The third experiment found that face masks that were similar to the targets had an effect equal to dissimilar ones. In all three experiments no interaction was found between the type of mask and the thresholds for the familiarity and naming criteria. The results are discussed in terms of information-processing models of face...

Estimating Skeletal Muscle Fascicle Curvature From B-Mode Ultrasound Image Sequences

We address the problem of tracking in vivo muscle fascicle shape and length changes using ultrasound video sequences. Quantifying fascicle behavior is required to improve understanding of the functional significance of a muscles geometric properties. Ultrasound imaging provides a noninvasive means of capturing information on fascicle behavior during dynamic movements; to date however, computational approaches to assess such images are limited. Our approach to the problem is novel because we permit fascicles to take up nonlinear shape configurations. We achieve this using a Bayesian tracking framework that is: 1) robust, conditioning shape estimates on the entire history of image observations; and 2) flexible, enforcing only a very weak Gaussian Process shape prior that requires fascicles to be locally smooth. The method allows us to track and quantify fascicle behavior in vivo during a range of movements, providing insight into dynamic changes in muscle geometric properties which may be linked to patterns of activation and intramuscular forces and pressures.

Automatic Wrinkle Detection Using Hybrid Hessian Filter

Aging as a natural phenomenon affects different parts of the human body under the influence of various biological and environmental factors. The most pronounced changes that occur on the face is the appearance of wrinkles, which are the focus of this research. Accurate wrinkle detection is an important task in face analysis. Some have been proposed in the literature, but the poor localization limits the performance of wrinkle detection. It will lead to false wrinkle detection and consequently affect the processes such as age estimation and clinician score assessment. Therefore, we propose a hybrid Hessian filter (HHF) to cope with the identified problem. HHF is composed of the directional gradient and Hessian matrix. The proposed filter is conceptually simple, however, it significantly increases the true wrinkle localization when compared with the conventional methods. In the experimental setup, three coders have been instructed to annotate the wrinkle of 2D forehead image manually. The inter-reliability among three coders is 93 % of Jaccard similarity index (JSI). In comparison to the state-of-the-art Cula method (CLM) and Frangi filter, HHF yielded the best result with a mean JSI of 75.67 %. We noticed that the proposed method is capable of detecting the medium to coarse wrinkle but not the fine wrinkle. Although there is a gap between human annotation and automated detection, this work demonstrates that HHF is a remarkably strong filter for wrinkle detection. From the experimental results, we believe that our findings are notable in terms of the JSI.

Semiautomated analysis of embryoscope images: Using localized variance of image intensity to detect embryo developmental stages.

Embryo selection in in vitro fertilization (IVF) treatment has traditionally been done manually using microscopy at intermittent time points during embryo development. Novel technique has made it possible to monitor embryos using time lapse for long periods of time and together with the reduced cost of data storage, this has opened the door to long‐term time‐lapse monitoring, and large amounts of image material is now routinely gathered. However, the analysis is still to a large extent performed manually, and images are mostly used as qualitative reference. To make full use of the increased amount of microscopic image material, (semi)automated computer‐aided tools are needed. An additional benefit of automation is the establishment of standardization tools for embryo selection and transfer, making decisions more transparent and less subjective. Another is the possibility to gather and analyze data in a high‐throughput manner, gathering data from multiple clinics and increasing our knowledge of early human embryo development. In this study, the extraction of data to automatically select and track spatio‐temporal events and features from sets of embryo images has been achieved using localized variance based on the distribution of image grey scale levels. A retrospective cohort study was performed using time‐lapse imaging data derived from 39 human embryos from seven couples, covering the time from fertilization up to 6.3 days. The profile of localized variance has been used to characterize syngamy, mitotic division and stages of cleavage, compaction, and blastocoel formation. Prior to analysis, focal plane and embryo location were automatically detected, limiting precomputational user interaction to a calibration step and usable for automatic detection of region of interest (ROI) regardless of the method of analysis. The results were validated against the opinion of clinical experts.