Showing papers by "William A. P. Smith published in 2013"
TL;DR: This paper starts by recovering 3D shape using a novel algorithm which incorporates generalization error of the model obtained from empirical measurements, and describes two methods to recover facial texture, diffuse lighting, specular reflectance, and camera properties from a single image.
Abstract: In this paper, we present a complete framework to inverse render faces with a 3D Morphable Model (3DMM). By decomposing the image formation process into geometric and photometric parts, we are able to state the problem as a multilinear system which can be solved accurately and efficiently. As we treat each contribution as independent, the objective function is convex in the parameters and a global solution is guaranteed. We start by recovering 3D shape using a novel algorithm which incorporates generalization error of the model obtained from empirical measurements. We then describe two methods to recover facial texture, diffuse lighting, specular reflectance, and camera properties from a single image. The methods make increasingly weak assumptions and can be solved in a linear fashion. We evaluate our findings on a publicly available database, where we are able to outperform an existing state-of-the-art algorithm. We demonstrate the usability of the recovered parameters in a recognition experiment conducted on the CMU-PIE database.
192 citations
TL;DR: This article serves as a guide to a variety of users, describing the functionality, performance and structure of DL_MESO, a parallel mesoscale simulation package capable of dissipative particle dynamics and the lattice Boltzmann equation method.
Abstract: DL_MESO is a parallel mesoscale simulation package capable of dissipative particle dynamics and the lattice Boltzmann equation method. It has been developed at Daresbury Laboratory for the United Kingdom Collaborative Computational Project known as CCP5. Capable of addressing industrially relevant tasks, but written to support academic research, it has a wide range of applications and scales to thousands of processors on high-performance computing platforms yet runs efficiently on smaller commodity clusters and single processor personal computers. This article serves as a guide to a variety of users, describing the functionality, performance and structure of this simulation package. Representative examples highlighting the capabilities of DL_MESO are given for each of the two methodologies available. Future directions for the package are discussed towards the end of the article.
137 citations
TL;DR: This paper presents a new database suitable for both 2-D and 3-D face recognition based on photometric stereo (PS): the Photoface database, collected using a custom-made four-source PS device designed to enable data capture with minimal interaction necessary from the subjects.
Abstract: This paper presents a new database suitable for both 2-D and 3-D face recognition based on photometric stereo (PS): the Photoface database. The database was collected using a custom-made four-source PS device designed to enable data capture with minimal interaction necessary from the subjects. The device, which automatically detects the presence of a subject using ultrasound, was placed at the entrance to a busy workplace and captured 1839 sessions of face images with natural pose and expression. This meant that the acquired data is more realistic for everyday use than existing databases and is, therefore, an invaluable test bed for state-of-the-art recognition algorithms. The paper also presents experiments of various face recognition and verification algorithms using the albedo, surface normals, and recovered depth maps. Finally, we have conducted experiments in order to demonstrate how different methods in the pipeline of PS (i.e., normal field computation and depth map reconstruction) affect recognition and verification performance. These experiments help to 1) demonstrate the usefulness of PS, and our device in particular, for minimal-interaction face recognition, and 2) highlight the optimal reconstruction and recognition algorithms for use with natural-expression PS data. The database can be downloaded from http://www.uwe.ac.uk/research/Photoface.
55 citations
29 Jul 2013
TL;DR: It is shown that the spherical AST outperforms planar AST applied to equirectangular images in terms of repeatability under rotation and the Accelerated Segment Test AST corner detector is extended.
Abstract: We extend the Accelerated Segment Test AST corner detector to operate on spherical images. We represent images using a discrete geodesic grid composed of triangular or hexagonal pixels. This representation has a number of advantages over the more common equirectangular parameterisation and, in the case of hexagonal pixels, leads more naturally to the discrete circular discs used in the AST. We present results on fully spherical imagery and show that our spherical AST outperforms planar AST applied to equirectangular images in terms of repeatability under rotation.
7 citations
30 Sep 2013
TL;DR: The RB formalism is described and the strategy for its implementation in DL_ POLY_4 formerly known as DL_POLY_3, which uses a very different form of parallelism, namely domain decomposition, is outlined.
Abstract: To remove problematic, high frequency degrees of freedom from a molecular model, modellers often use rigid body dynamics. The method also has additional benefits, for example it allows molecular charge distributions to be conveniently represented by multipolar moments. Rigid bodies are a well established feature within DL_POLY_ Classic formerly known as DL_POLY_2, which employs the replicated data parallelisation methodology. This paper briefly describes the RB formalism and outlines the strategy for its implementation in DL_POLY_4 formerly known as DL_POLY_3, which uses a very different form of parallelism, namely domain decomposition.