19th Ave New York, NY 95822, USA

# 计算机代写|计算机视觉代写Computer Vision代考|CS5670 Vision as a set of inverse problems. Regularization

my-assignmentexpert™提供最专业的一站式服务：Essay代写，Dissertation代写，Assignment代写，Paper代写，Proposal代写，Proposal代写，Literature Review代写，Online Course，Exam代考等等。my-assignmentexpert™专注为留学生提供Essay代写服务，拥有各个专业的博硕教师团队帮您代写，免费修改及辅导，保证成果完成的效率和质量。同时有多家检测平台帐号，包括Turnitin高级账户，检测论文不会留痕，写好后检测修改，放心可靠，经得起任何考验！

## 计算机代写|计算机视觉代写Computer Vision代考|Vision as a set of inverse problems. Regularization

Many goals of vision can be described as “inverse problems,” because inferences from data are needed to derive models of the world properties that projected those image data. Such inverse problems are formally ill-posed. A powerful way to overcome the inherently ill-posed nature of many visual tasks is to re-formulate, or regularise them as optimisation problems, which can be approached essentially as problems of calculus because the solution to the task corresponds to finding an extremum. The method involves specifying some functional (typically an integral operator) consisting of two terms:

1. a “data” term, in which image parameters are represented;
2. a “smoothness” (or regularisation) term, embodying world constraints.
These two terms are combined by some weighting parameter $\lambda$ specifying their relative importance, i.e. the degree to which either the data dominates, or the background assumptions dominate, the solution.

The goal of detecting faces and recognizing their identity has long been one of the “Holy Grail” problems in computer vision. It is a hard problem for all of the reasons we have encountered that generally make computer vision hard:

• Faces are surfaces on 3D objects (heads). Therefore the images they project depend on the perspective angle between object and camera, the rotation of the object around its own axes, and the illuminant.
• Facial surfaces have relief, and so parts (e.g. noses) can occlude other parts. Hair can also create random occlusions and shadows.
• Surface relief causes shading and shadows to depend upon the angle of the illuminant, and whether it is an extended or a point source.
• Faces have variable specularity (dry skin may be Lambertian, oily or sweaty skin may be specular). As always, this confounds the interpretation of the reflectance map.
• Parts of faces can move around relative to other parts (eye movements; lip movements; eyebrows and winks).
• Humans put things on their faces (e.g. glasses, cosmetics, cigarettes) and change their facial hair (moustaches, eyebrows). They also use their faces as organs of expression, and so the surface isn’t even rigid. (Ideally one would like not only to be able to detect and recognize faces, but also to interpret and classify their expressions.)

## 计算机代写|计算机视觉代写Computer Vision代考|Issues in detecting, recognising, and interpreting faces

As usual, this domain of computer vision raises questions such as:

1. What is the best representation to use for faces?
2. Must this be treated as a 3D (object-based) or 2D (image-based) problem?
3. How can invariances to size (hence distance), location, pose, and angle of view be achieved? (A face should acquire the same representation under such transformations, for matching purposes.)
4. What are the generic (i.e. universal) properties of all faces that we can rely upon, in order to reliably detect the presence of a face?
5. What are the particular features that we can rely upon to recognize the identity of any given face?
6. What is the best way to handle “integration of evidence,” and incomplete information, and to make decisions under uncertainty?
7. How can we handle the transformations that can occur in a given person’s face, either through natural, or unnatural means?

## 计算机代写|计算机视觉代写COMPUTER VISION代考|VISION AS A SET OF INVERSE PROBLEMS. REGULARIZATION

1. “数据”项，其中表示图像参数；
2.一个“平滑度”orregularisation术语，体现了世界的约束。 这两项由一些加权参数组合 $\lambda$ 指定它们的相对重要性，即数据主导或背景假设主导解决方案的程度。
长期以来，检测人脸并识别其身份的目标一直是计算机视觉领域的“圣杯”问题之一。由于我们遇到的所有通常使计算机视觉困难的原因，这是一 个难题:
• 面部表面有浮雕等部位e.g.noses 可以遮挡其他部分。头发还可以产生随机遮挡和阴影。
• 表面起伏导致阴影和阴影取决于光源的角度，以及它是扩展光源还是点光源。
• 面孔具有可变的镜面反射dryskinmaybeLambertian, oilyorsweatyskinmaybespecular.一如既往，这混淆了反射图的解释。
• 面部的一部分可以相对于其他部分四处移动eyemovements; lipmovements; eyebrowsandwinks.
• 人类把东西放在脸上e.g.glasses, cosmetics, cigarettes并改变他们的面部毛发moustaches, eyebrows. 他们也用他们的脸作为表达器 官，所以表面甚至不是僵硬的。
Ideallyonewouldlikenotonlytobeabletodetectandrecognize faces, butalsotointerpretandclassi fytheirexpressions.

## 计算机代写|计算机视觉代写COMPUTER VISION代 考|ISSUES IN DETECTING, RECOGNISING, AND INTERPRETING FACES

1. 用于人脸的最佳表示是什么?
2. 必须将其视为 3Dobject – based或二维 image – based问题?
3. 尺寸不变性如何hencedistance、位置、姿势、视角如何实现?
Afaceshouldacquirethesamerepresentationundersuchtransformations, formatchingpurposes.
4. 通用的有哪些i.e. universal我们可以依赖的所有面部的属性，以便可靠地检测面部的存在?
5. 我们可以依靠哪些特定特征来识别任何给定面孔的身份?
6. 处理“证据整合”和不完整信息以及在不确定情况下做出决策的最佳方式是什么?
7. 我们如何处理通过自然或非自然方式发生在给定人脸上的变化?

## Matlab代写

MATLAB 是一种用于技术计算的高性能语言。它将计算、可视化和编程集成在一个易于使用的环境中，其中问题和解决方案以熟悉的数学符号表示。典型用途包括：数学和计算算法开发建模、仿真和原型制作数据分析、探索和可视化科学和工程图形应用程序开发，包括图形用户界面构建MATLAB 是一个交互式系统，其基本数据元素是一个不需要维度的数组。这使您可以解决许多技术计算问题，尤其是那些具有矩阵和向量公式的问题，而只需用 C 或 Fortran 等标量非交互式语言编写程序所需的时间的一小部分。MATLAB 名称代表矩阵实验室。MATLAB 最初的编写目的是提供对由 LINPACK 和 EISPACK 项目开发的矩阵软件的轻松访问，这两个项目共同代表了矩阵计算软件的最新技术。MATLAB 经过多年的发展，得到了许多用户的投入。在大学环境中，它是数学、工程和科学入门和高级课程的标准教学工具。在工业领域，MATLAB 是高效研究、开发和分析的首选工具。MATLAB 具有一系列称为工具箱的特定于应用程序的解决方案。对于大多数 MATLAB 用户来说非常重要，工具箱允许您学习应用专业技术。工具箱是 MATLAB 函数（M 文件）的综合集合，可扩展 MATLAB 环境以解决特定类别的问题。可用工具箱的领域包括信号处理、控制系统、神经网络、模糊逻辑、小波、仿真等。