Showing papers by "William A. P. Smith published in 2007"

Molecular dynamics using atomic-resolution structure reveal structural fluctuations that may lead to polymerization of human Cu–Zn superoxide dismutase

Abstract: Mutations of the gene encoding Cu-Zn superoxide dismutase (SOD1) cause 20% of the familial cases of the progressive neurodegenerative disease ALS A growing body of evidence suggests that in familial ALS (FALS) it is the molecular behavior of the metal-depleted SOD1 dimer that leads to a gain of toxic properties by misfolding, unfolding, and aggregation Structural studies have so far provided static snapshots on the behavior of the wild-type enzyme and some of the FALS mutants New approaches are required to map out the structural trajectories of the molecule Here, using our 115-A resolution structure of fully metallated human SOD1 and highly parallelized molecular dynamics code on a high-performance capability computer, we have undertaken molecular dynamics calculations to 4,000 ps to reveal the first stages of misfolding caused by metal deletion Large spatial and temporal fluctuations of the "electrostatic" and "Zn-binding" loops adjacent to the metal-binding sites are observed in the apo-enzyme relative to the fully metallated dimer These early misfolding events expose the beta-barrels of the dimer to the external environment, allowing close interactions with adjacent molecules Protection of the beta-edge of the protein can be partially restored by incorporating a single Zn molecule per dimer These calculations reveal an essential step in the formation of the experimentally observed self-aggregations of metal-depleted FALS mutant SOD1 This result also has implications for the role of demetallated wild-type SOD1 in sporadic cases of ALS, for which the molecular cause still remains undiscovered

A Coupled Statistical Model for Face Shape Recovery From Brightness Images

Abstract: We focus on the problem of developing a coupled statistical model that can be used to recover facial shape from brightness images of faces. We study three alternative representations for facial shape. These are the surface height function, the surface gradient, and a Fourier basis representation. We jointly capture variations in intensity and the surface shape representations using a coupled statistical model. The model is constructed by performing principal components analysis on sets of parameters describing the contents of the intensity images and the facial shape representations. By fitting the coupled model to intensity data, facial shape is implicitly recovered from the shape parameters. Experiments show that the coupled model is able to generate accurate shape from out-of-training-sample intensity images

Gender Classification using Shape from Shading

Abstract: The aim in this paper is to show how to use the 2.5D facial surface normals (needle-maps) recovered using shape from shading (SFS) to improve the performance of gender classification. We incorporate principal geodesic analysis (PGA) into SFS to guarantee the recovered needle-maps is a possible example defined by a statistical model. Because the recovered facial needlemaps satisfy data-closeness constraint, they not only give the facial shape information, but also combine the image intensity implicitly. Experiments show that this combination gives better gender classification performance than using facial shape or texture information alone.

Gender Classification using Shape from Shading

Abstract: The aim in this paper is to show how to use the 2.5D facial surface normals (needle-maps) recovered using shape from shading (SFS) to improve the performance of gender classification. We incorporate principal geodesic analysis (PGA) into SFS to guarantee the recovered needle-maps is a possible example defined by a statistical model. Because the recovered facial needlemaps satisfy data-closeness constraint, they not only give the facial shape information, but also combine the image intensity implicitly. Experiments show that this combination gives better gender classification performance than using facial shape or texture information alone.

Weighted principal geodesic analysis for facial gender classification

Abstract: In this paper, we describe a weighted principal geodesic analysis (WPGA) method to extract features for gender classification based on 2.5D facial surface normals (needle-maps) which can be extracted from 2D intensity images using shape-from-shading (SFS). By incorporating the weight matrix into principal geodesic analysis (PGA), we control the obtained principal axis to be in the direction of the variance on gender information. Experiments show that using WPGA, the leading eigenvectors encode more gender discriminating power than using PGA, and that gender classification based on leading WPGA parameters is more accurate and stable than based on leading PGA parameters.

Use of massively parallel molecular dynamics simulations for radiation damage in pyrochlores

Abstract: DL_POLY_3 is a general purpose molecular dynamics (MD) simulation package designed to simulate systems of the order of tens of millions of particles and beyond by efficiently harnessing the power of modern computer clusters. Here we discuss the package design, functionality and report on performance and capability limits. We then report the application of DL_POLY_3 to study radiation cascades in Gd2Ti2O7 and Gd2Zr2O7, potential materials for high-level radioactive waste storage and discuss problems associated with the analysis of the cascades. We see little direct amorphisation but rather the start of a transition to the fluorite structure which is more pronounced for the Zr than the Ti compound.

Learning Mixture Models for Gender Classification Based on Facial Surface Normals

Abstract: The aim in this paper is to show how to discriminate gender using a parameterized representation of fields of facial surface normals (needle-maps) We make use of principle geodesic analysis (PGA) to parameterize the facial needle-maps Using feature selection, we determine the selected feature set which gives the best result in distinguishing gender Using the EM algorithm we distinguish gender by fitting a two component mixture model to the vectors of selected features Results on real-world data reveal that the method gives accurate gender discrimination results

A new framework for grayscale and colour non-lambertian shape-from-shading

Abstract: In this paper we show how arbitrary surface reflectance properties can be incorporated into a shape-from-shading scheme, by using a Riemannian minimisation scheme to minimise the brightness error. We show that for face images an additional regularising constraint on the surface height function is all that is required to recover accurate face shape from single images, the only assumption being of a single light source of known direction. The method extends naturally to colour images, which add additional constraints to the problem. For our experimental evaluation we incorporate the Torrance and Sparrow surface reflectance model into our scheme and show how to solve for its parameters in conjunction with recovering a face shape estimate. We demonstrate that the method provides a realistic route to non-Lambertian shape-from-shading for both grayscale and colour face images.

Face Recognition Using Principal Geodesic Analysis and Manifold Learning

Abstract: This paper describes how face recognition can be effected using 3D shape information extracted from single 2D image views. We characterise the shape of the field of facial normals using a statistical model based on principal geodesic analysis. The model can be fitted to 2D brightness images of faces to recover a vector of shape parameters. Since it captures variations in a field of surface normals, the dimensionality of the shape vector is twice the number of image pixels. We investigate how to perform face recognition using the output of PGA by applying a number of dimensionality reduction techniques including principal components analysis, locally linear embedding, locality preserving projection and Isomap.

Facial Shape Spaces from Surface Normals and Geodesic Distance

Abstract: The analysis of shape-variations due to changes in facial expression and gender difference is a key problem in face recognition. In this paper, we draw on ideas from the field of statistical shape analysis to construct shape-spaces that span facial expressions and gender, and use the resulting shape-model to perform face recognition under varying expression and gender. Our novel contribution is to show how to construct shape-spaces over fields of surface normals rather than Cartesian landmark points. According to this model face needle-maps (or fields of surface normals) are points in a high-dimensional manifold referred to as a shape-space. The similarity between faces and gender difference is measured using a number of alternative geodesic, Euclidean and cosine distance between points on the manifold. In a recognition experiment we compare the perfomance distance with Euclidean, cosine and geodesic distance associated with the shape manifold. Here we explore if the distances used to distinguish gender and recognise the same for under different expressions.

Face recognition with irregular region spin images

Abstract: This paper explores how spin images can be constructed using shapefrom-shading information and used for the purpose of face recognition. We commence by extracting needle maps from gray-scale images of faces, using a mean needle map to enforce the correct pattern of facial convexity and concavity. Spin images [6] are estimated from the needle maps using local spherical geometry to approximate the facial surface. Our representation is based on spin image histograms for an arrangement of image patches. Comparing to our previous spin image approach, the current one has two basic difference: Euclidean distance is replaced by geodesic distance; Irregular face region is applied to better fit face contour. We demonstrate how this representation can be used to perform face recognition across different subjects and illumination conditions. Experiments show the method to be reliable and accurate, and the recognition precision reaches 93% on CMU PIE sub-database.

Facial shape-from-shading using principal geodesic analysis and robust statistics

Abstract: In this paper we make two contributions to the problem of recovering surface shape from single images of faces. The first of these is to develop a representation of the distribution of surface normals based on the exponential map, and to show how to model shape-deformations using principal geodesic analysis on the exponential map. The second contribution is to show how ideas from robust statistics can be used to fit the model to facial images in which there is significant self-shadowing. The method is evaluated on both synthetic and real-world images. It is demonstrated to effectively fill-in the facial surface when more than 30% of the area is subject to self-shadowing.

Statistical methods for surface integration

Abstract: In this paper we show how statistical constraints can be incorporated into the surface integration process. This problem aims to reconstruct the surface height function from a noisy field of surface normals. We propose two methods that employ a statistical model that captures variations in surface height. The first uses a coupled model that captures the variation in a training set of face surfaces in both the surface normal and surface height domain. The second is based on finding the parameters of a surface height model directly from a field of surface normals. We present experiments on ground truth face data and compare the results of the two methods with an existing surface integration technique.

Surface Integration: Two Statistical Approaches

Abstract: In this paper we show how statistical constraints can be incorporated into the surface integration process. This problem aims to reconstruct the surface height function from a noisy field of surface normals. We propose two methods that employ a statistical model that captures variations in surface height. The first uses a coupled model that captures the variation in a training set efface surfaces in both the surface normal and surface height domain. The second is based on finding the parameters of a surface height model directly from afield of surface normals. We present experiments on ground truth face data and compare the results of the two methods with an existing surface integration technique.

Testing viewpoint invariance in the neural representation of faces: an MEG study

Abstract: The aim of this study was to determine the extent to which the neural representation of faces in the visual cortex is viewpoint invariant. MEG was used to measure evoked responses to faces during an adaptation paradigm. Using familiar and unfamiliar faces, we compared the amplitude of the M170 response to repeated images of the same face compared to images of different faces. We found a reduction in the M170 amplitude to repeated presentations of the same face image compared to images of different faces when shown from the same viewpoint. To establish if this adaptation to the identity of a face was invariant to changes in viewpoint, we varied the viewing angle of the face within a block. In order to exert strict control over the viewpoint from which the face was viewed, we used 3D models recovered from single images using shape-from-shading. This makes the study unique in its use of techniques from machine vision in order to test human visual processes. We found a reduction in response was no longer evident when images of the same face were shown from different viewpoints. These results imply that the face-selective M170 response either reflects an early stage of face processing or that the computations underlying face recognition depend on a viewpoint-dependent neuronal representation.