Dissertations / Theses on the topic 'Rendering faces'

In this thesis, we present a complete framework to inverse render faces with a 3D Morphable Model. By decomposing the image formation process into a geometric and photometric part, 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 globally optimal solution can be found. We start by recovering 3D shape using a novel algorithm which incorporates generalisation errors of the model obtained from empirical measurements. The algorithm is extended so it can efficiently deal with mixture distributions. We then describe three methods to recover facial texture, and for the second and third, diffuse lighting, specular reflectance and camera properties from a single image. These methods make increasingly weak assumptions and can all be solved in a linear fashion. We further modify our framework so it accounts for global illumination effects. This is achieved by incorporating statistical models for ambient occlusion and bent normals into the image formation model. We show that solving for ambient occlusion and bent normal parameters as part of the fitting process improves the accuracy of the estimated texture map and illumination environment. We present results on challenging data, rendered under complex natural illumination with both specular reflectance and occlusion of the illumination environment. We evaluate our findings on publicly available datasets, where we are able to obtain state-of-the-art results. Finally, we present a practical method to synthesise a larger population from a small training-set and show how the new instances can be used to build a flexible PCA model.

Human viewers are extremely sensitive to the appearanceof peoples faces, which makes the rendering of realistic human faces a challenging problem. Techniques for doing this have continuously been invented and evolved since more than thirty years.

This thesis makes use of recent methods within the area of image based rendering, namely the acquisition of reflectance fields from human faces. The reflectance fields are used to synthesize and realistically render models of human faces.

A shape from shading technique, assuming that human skin adheres to the Phong model, has been used to estimate surface normals. Belief propagation in graphs has then been used to enforce integrability before reconstructing the surfaces. Finally, the additivity of light has been used to realistically render the models.

The resulting models closely resemble the subjects from which they were created, and can realistically be rendered from novel directions in any illumination environment.

The human face is a fascinating, but extremely complex object; the research project described is concerned with the computer generation and animation of faces. However, the age old captivation with the face transforms into a major obstacle when creating synthetic faces. The face and head are the most visible attributes of a person. We master the skills of recognising faces and interpreting facial movement at a very early age. As a result, we are likely to notice the smallest deviation from our concept of how a face should appear and behave. Computer animation in general, is often perceived to be ``wooden' and very ``rigid'; the aim is therefore to provide facilities for the generation of believable faces and convincing facial movement. The major issues addressed within the project concern the modelling of a large variety of faces and their animation. Computer modelling of arbitrary faces is an area that has received relatively little attention in comparison with the animation of faces. Another problem that has been considered is that of providing the user with adequate and effective control over the modelling and animation of the face. The Facial Animation, Construction and Editing System or FACES was conceived as a system for investigating these issues. A promising approach is to look a little deeper than the surface of the skin. A three-layer anatomical model of the head, which incorporates bone, muscle, skin and surface features, has been developed. As well as serving as a foundation which integrates all the facilities available within FACES, the advantage of the model is that it allows differing strategies to be used for modelling and animation. FACES is an interactive system, which helps with both the generation and animation of faces, while hiding the structural complexities of the face from the user. The software consists of four sub-systems; CONSTRUCT and MODIFY cater for modelling functionality, while ANIMATE allows animation sequences to be generated and RENDER provides for shading and motion evaluation.

Thesis (M. Eng.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2006.
Includes bibliographical references (p. 59-60).
Human skin exhibits complex light reflectance properties that make it difficult to render realistically. In recent years, many techniques have been introduced to render skin, with varying degrees of complexity and realism. In this thesis, I will implement several of these techniques, and use them to render scenes with various lighting and geometry parameters, in order to compare their strengths and weaknesses. My goal is to provide a clearer understanding of which rendering techniques are most effective in different scenarios.
by Arturo Andrew Arizpe.
M.Eng.

In volumetric visualization, the body centered cubic grid (BCC) and its reciprocal, the face centered cubic grid (FCC), are despite their good sampling properties  not well off regarding available rendering software and tools. Described in this thesis is the development of an extension for the volume rendering engine Voreen, implementing two recently presented  GPU accelerated reconstruction algorithms for these grids, along with a simple nearest neighbor method. These reconstruction  methods replaces the trilinear reconstruction  method used for data stored in a Cartesian cubic grid (CC). The goal is for the produced software to be useful for efficiently visualizing results from experiments with the BCC and FCC grids and thus help make such data easier to observe. The performance and rendering quality of the new raycasters is measured and compared to Voreen's existing Cartesian cubic ray caster. The experimental results show that the raycasters can render data in the BCC and FCC format at interactive frame rates while maintaining comparable visual quality.

The body centered cubic grid (BCC) and face centered cubic grid (FCC) offer improved sampling properties when compared to the cartesian grid. Despite this there is little software and hardware support for volume rendering of data stored in one of these grids. This project is a continuation of a project adding support for such grids to the volume rendering engine Voreen. This project has three aims. Firstly, to implement new interpolation methods capable of rendering at interactive frame rates. Secondly, to improve the software by adding an alternate volume storage format offering improved frame rates for BCC methods. And thirdly, because of the issues when comparing image quality between different grid types due to aliasing, to implement a method unbiased in terms of post-aliasing. The existing methods are compared to the newly implemented ones in terms of frame rate and image quality and the results show that the new volume format improve the frame rate significantly, that the new BCC interpolation method offers similar image quality at better performance compared to existing methods and that the unbiased method produces images of good quality at the expense of speed.

The accurate modeling of the variability of illumination in a class of images is a fundamental problem that occurs in many areas of computer vision and graphics. For instance, in computer vision there is the problem of facial recognition. Simply, one would hope to be able to identify a known face under any illumination. On the other hand, in graphics one could imagine a system that, given an image, the illumination model could be identified and then used to create new images. In this thesis we describe a method for learning the illumination model for a class of images. Once the model is learnt it is then used to render new images of the same class under the new illumination. Results are shown for both synthetic and real images. The key contribution of this work is that images of known objects can be re-illuminated using small patches of image data and relatively simple kernel regression models. Additionally, our approach does not require any knowledge of the geometry of the class of objects under consideration making it relatively straightforward to implement. As part of this work we will examine existing geometric and image-based re-lighting techniques; give a detailed description of our geometry-free face re-lighting process; present non-linear regression and basis selection with respect to image synthesis; discuss system limitations; and look at possible extensions and future work.
M.S.
School of Electrical Engineering and Computer Science
Engineering and Computer Science
Computer Science MS

Scanning people and generating realistic face models is becoming nowadays an important branch in Computer Graphics. The main applications are the creation of 3D selfies that can be immediately viewed or printed in 3D, the creation of 3D avatars for augmented or virtual reality applications, content creation for games or movies, or the developing of 3D face authentication algorithms (more robust respect to the 2D counterpart). This thesis provides an overview on a common issue that arises when scanning people wearing glasses and in turn proposes different solutions to solve or at least bypass the original problem, showing the strengths and the weaknesses. Distinct techniques for automatically putting a pair of glasses on a human face are also present. Furthermore, a study on the physical properties of glasses is documented, involving both the frame and the lenses, along with a brief description of the Blender software and its main features that were essential during the development.

Thesis (M.S.)--Florida State University, 2004.
Advisor: Dr. David C. Banks, Florida State University, College of Arts and Sciences, Dept. of Computer Science. Title and description from dissertation home page (viewed Oct. 27, 2004). Includes bibliographical references.

The first approach, described in the first part of this thesis, studies 3D face modeling from multi-views. Today human face modeling and animation techniques are widely used to generate virtual characters and models. Such characters and models are used in movies, computer games, advertising, news broadcasting and other activities. We propose an efficient method to estimate the poses, the global shape and the local structures of a human head recorded in multiple face images or a video sequence by using a generic wireframe face model. Based on this newly proposed method, we have successfully developed a pose invariant face recognition system and a pose invariant face contour extraction method.
The objective of this thesis is to model and render complex scenes or objects from multiple images taken from different viewpoints. Two approaches to achieve this objective were investigated in this thesis. The first one is for known objects with prior geometrical models, which can be deformed to match the objects recorded in multiple input images. The second one is for general scenes or objects without prior geometrical models.
The proposed algorithms in this thesis were tested on many real and synthetic data. The experimental results illustrate their efficiency and limitations.
The second approach, described in the second part of this thesis, investigates 3D modeling and rendering for general complex scenes. The entertainment industry touches hundreds of millions of people every day, and synthetic pictures and 3D reconstruction of real scenes, often mixed with actual film footage, are now common place in computer games, sports broadcasting, TV advertising and feature films. A series of techniques has been developed to complete this task. First, a new view-ordering algorithm was proposed to organize and order an unorganized image database. Second, a novel and efficient multiview feature matching approach was developed to calibrate and track all views. Finally, both match propagation based and Bayesian based methods were developed to produce 3D scene models for rendering.
Yao Jian.
"September 2006."
Adviser: Wai-Kuen Chan.
Source: Dissertation Abstracts International, Volume: 68-03, Section: B, page: 1849.
Thesis (Ph.D.)--Chinese University of Hong Kong, 2006.
Includes bibliographical references (p. 170-181).
Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web.
Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web.
Abstracts in English and Chinese.
School code: 1307.

碩士
國立彰化師範大學
電機工程學系
99
This thesis presents stereo vision used to reconstruct a 3D face model. The main method uses the facial features to find the corresponding points in left and right images. The disparity relation of the corresponding left and right image points is used to calculate the face depth information. This information is used to reconstruct a three-dimensional face model. The feature point demarcation is based on principle component analysis (PCA) and active shape model (ASM) to complete face image segmentation and the corresponding points search. The general camera lens has radial distortion, which impacts the matching errors on the left and right images. The camera calibration method is used to calculate the internal and external camera parameters for image distortion correction. The left and right image disparity map is used with Matlab to draw a three-dimensional face model. The experimental results demonstrate the reconstructed 3D face model has good identification properties.

M.A.
Child migration in society today appears to be escalating at a very high rate. The influx of child migrants has put an enormous strain to social workers in terms of rendering family preservation and reunification services. In most cases families are not available to enable social workers to render family preservation and reunification services. International Social Services (ISS) assist in tracing families of across the border child migrants but the service does not extend beyond reunifying the child with family. There are no monitoring and tracking systems to prevent the child from migrating again. The aim of this study is to identify and describe the challenges faced by social workers in rendering family preservation and reunification services to child migrants. The objectives of the study are as follows: to conduct a literature study on family preservation and reunification services and the child migrant; to determine social workers’ strategies used in family preservation and reunification services to child migrants, and the challenges they are experiencing; to obtain social workers’ views on effective strategies that can be implemented in rendering family preservation and reunification services to child migrants and to make recommendations that will assist social workers in rendering family preservation and reunification services. A qualitative study of two focus groups consisting of 15 social workers each from Department of Social Development Johannesburg Region was conducted. The research findings revealed the need of coming up with strategies for family preservation and reunification services to child migrants. Presently social workers have no strategies but processes only.

The study aimed to develop an in-depth understanding of the challenges faced by social workers rendering Occupational Social Work (OSW) services, particularly in establishing and maintaining workplace support groups for employees living with HIV and AIDS. The study used a qualitative research approach, aided by exploratory, descriptive and contextual research design. Data was collected from eight participants with semi-structured interviews and was analyzed by applying eight steps for data analysis as proposed by Tesch, cited in Creswell (2003:191). Data verification was done by applying Guba’s model (in Krefting, 1991). The themes, which emerged from the interviews, were: the recruitment of members in workplace support groups and the challenges faced by social workers rendering OSW services. The findings indicate that, critical as their roles may be, social workers experience numerous challenges when providing workplace HIV and AIDS support groups. These include HIV and AIDS disclosure, management, organisational and logistical challenges
Social Work

In this thesis, studies on three problems in imaging systems will be discussed.

The first problem deals with re-rendering segments of online indoor room images with preferred textures through websites to try new decoration ideas. Previous methods need too much manual positioning and alignment. In the thesis, a novel approach is presented to automatically achieve a natural outcome with respect to indoor room geometry layout.

For the second problem, the laser electrophotographic system is eagerly looking for a digital halftoning algorithm that can deal with unequal printing resolution, since most halftoning algorithms are focused on equal resolution. In the thesis, a novel monochrome halftoning algorithm is presented to render continuous tone images with limited numbers of tone levels for laser printers with unequal printing resolution.

For the third problem, a novel face set recognition method is presented. Face set recognition is important for face video analysis and face clustering in multiple imaging systems. And it is very challenging considering the variation of image sharpness, face directions and illuminations for different frames, as well as the number and the order of images in the face set. To tackle the problem, a novel convolutional neural network system is presented to generate a fixed-dimensional compact feature representation for the face set. The system collects information from all the images in the set while having emphasis on more frontal and sharper face images, and it is regardless of the number and the order of images. The generated feature representations allow direct, immediate similarity computation for face sets, thus can be directly used for recognition. The experiment result shows that our method outperforms other state of-the-art methods on the public test dataset.