Feature Detection and Simplification of 3D Face Data with Facial Expressions

We propose an efficient framework to realistically render 3D faces with a reduced set of points. First, a robust active appearance model is presented to detect facial features in the projected faces under different illumination conditions. Then, an adaptive simplification of 3D faces is proposed to reduce the number of points, yet preserve the detected facial features. Finally, the point model is rendered directly, without such additional processing as parameterization of skin texture. This fully automatic framework is very effective in rendering massive facial data on mobile devices.     

人脸显性特征的融合构造方法及识别

 目前的人脸识别研究中,面部几何特征没有得到很好的利用.本文阐述了几何特征对于人脸识别的重要性,在此基础上提出了一种提取面部几何特征的新方法;通过融合几何信息和纹理信息构造出一种面部显性特征,并给出了相应的人脸识别方法.这种新的人脸识别方法相对于基于统计学习的子空间方法具有一定的优势,同时也可作为后者的有益补充.实验表明,本文提出的人脸表示特征及识别方法对人脸表情变化和环境光照变化均有一定的鲁棒性.     

Improving Deep Learning Feature with Facial Texture Feature for Face Recognition

Face recognition in the reality, is a challenging problem, due to varieties in illumination, background, pose etc. Recently, the deep learning based face recognition algorithm is able to learn effective face features to obtain a very impressive performance. However, this kind of face recognition algorithm completely relies on the machine learning based face features, while ignores the useful experience in hand-craft features which have been studied in a long period. Therefore, a face recognition based on facial texture feature aided deep learning feature (FTFA-DLF) is proposed in this paper. The proposed FTFA-DLF is able to combine the benefits of deep learning and hand-craft features. In the proposed FTFA-DLF method, the hand-craft features are texture features extracted from the eyes, nose, and mouth regions. Then, the hand-craft features are used to aid deep learning features by adding both deep learning and hand-craft features into the objective function layer, which adaptively adjusts the deep learning features so that it can better cooperate with the hand-craft features and obtain a better face recognition performance. Experimental results show that the proposed face recognition algorithm on the LFW face database to achieve the accuracy rate of 97.02%.     

融合面部表情和肢体动作特征的情绪识别

为了从视频数据判断人的情绪,首先提取视频数据中的时空特征,并用其表征情绪特征,然后分别用典型相关分析算法和稀疏保持典型相关分析算法融合面部情绪特征和肢体动作情绪特征,最后用最近邻分类和支持向量机分类分别对情绪分类识别.实验结果表明,稀疏保持典型相关分析融合算法优于典型相关分析融合算法,能得到90.48％的情绪识别率.     

基于统计学习优化SIFT的面部遮挡人脸识别

针对传统的人脸识别算法受面部遮挡的影响导致很难兼顾鲁棒性和保持原始图像核心信息的问题,本文提出了一种基于统计学习优化尺度不变特征变换的面部遮挡人脸识别算法。首先,利用SIFT将所有给定训练图像用一组局部特征描述符表示出来;然后,通过执行统计学习获得正常脸部图像SIFT特征的概率分布函数,利用获得的概率分布函数在新观察到的测试图像中检测异常SIFT特征;最后,计算测试图像与训练图像之间的相似度,并利用K近邻分类器完成人脸识别。在AR人脸数据库上的实验验证了本文算法的有效性及可靠性,实验结果表明,相比其它几种较为先进的人脸识别算法,本文算法取得了更强的识别鲁棒性。     

Machine Interpretation of Facial Expressions

Facial expressions are an important source of information for human interaction. Therefore, it would be desirable if computers were able to use this information to interact more naturally with the user. However, facial expressions are not always unambiguously interpreted even by competent humans. Consequently, soft computing techniques in which interpretations are given some belief value would seem appropriate. This paper describes how the mass assignment approach to constructing fuzzy sets from probability distributions has been applied to the low-level classification of pixels into facial feature classes based on their colour. It will also describe how similar approaches can be used for the analysis of facial expressions themselves.     

多姿态、光照不均人脸图像的子空间人脸识别方法研究

针对人脸识别中存在的光照不均匀问题,提出了一种预处理链技术,能达到很好的光照补偿效果。为了提高多姿态、多表情、多细节人脸图像的人脸识别率,设计了一种将最近邻分类器与支持向量机相结合的分类算法（NN-SVM）,基于该分类算法提出了一种基于Gabor变换和NN-SVM的子空间人脸识别方法。在FERET和ORL两大人脸数据库中对所提方法进行性能评估,实验结果表明所提出方法能有效的解决人脸识别中光照不均匀问题,大大的提高人脸识别率,而且相比其他现存的人脸识别方法,所设计的方法具有更好、更稳定的识别效果。     

融合人脸局部区域的表情识别

本文提出了一种融合人脸各个局部区域的表情识别方法,首先提取眼睛、鼻子和嘴巴三个局部区域,并对这三个局部区域和整个人脸区域分别进行表情识别,得到四个独立的识别结果,再利用融合策略对四个结果进行决策融合,所提方法在日本女性表情数据库JAFFE库和五邑大学人脸表情数据库CED_ WYU库上进行了广泛的实验,结果表明,通过融合人脸各局部区域的识别结果,表情识别率可以得到显著提高.     

基于不同Margin的人脸特征选择及识别方法

Margin在机器学习中具有很重要的意义,基于margin的特征选择方法就是从分类的角度对特征集各特征的权重进行分析。该文对不同的margin进行了分析,提出将sample-margin和hypothesis-margin分别作为特征选择标准对SBS特征选择方法进行改进,然后设计具有最佳超参数的SVM多项式分类器进行人脸识别。实验在FRERT人脸图像库上进行并与Relief特征选择方法进行了比较,对SVM和NN分类器的实验结果也进行了分析。实验结果显示：该文提出的人脸识别特征选择及识别方法是有效、适用的。     

Correlation Pattern Recognition for Face Recognition

Two-dimensional (2-D) face recognition (FR) is of interest in many verification (1:1 matching) and identification (1:N matching) applications because of its nonintrusive nature and because digital cameras are becoming ubiquitous. However, the performance of 2-D FR systems can be degraded by natural factors such as expressions, illuminations, pose, and aging. Several FR algorithms have been proposed to deal with the resulting appearance variability. However, most of these methods employ features derived in the image or the space domain whereas there are benefits to working in the spatial frequency domain (i.e., the 2-D Fourier transforms of the images). These benefits include shift-invariance, graceful degradation, and closed-form solutions. We discuss the use of spatial frequency domain methods (also known as correlation filters or correlation pattern recognition) for FR and illustrate the advantages. However, correlation filters can be computationally demanding due to the need for computing 2-D Fourier transforms and may not match well for large-scale FR problems such as in the Face Recognition Grand Challenge (FRGC) phase-II experiments that require the computation of millions of similarity metrics. We will discuss a new method [called the class-dependence feature analysis (CFA)] that reduces the computational complexity of correlation pattern recognition and show the results of applying CFA to the FRGC phase-II data     

人脸识别技术实现产业化 汉王推“人脸通”F710

通信技术用于门禁系统,除了NFC（近距离通信）技术以外,还有指纹系统,最近汉王科技推出的“人脸通”系统,使门禁系统迎来了“人脸”时代。     

抗年龄干扰的人脸识别

本文将人脸图像特征分解为身份特征部分和年龄干扰特征部分,并分别投影到两个独立的子空间.为了提高由字典张成的特征子空间的表征能力和区分能力,字典学习过程中同时引入了人脸图像的重构误差约束项和类别约束项.因此,任意的人脸特征都可以由学习到的身份字典和年龄字典投影到对应的身份子空间和年龄子空间,然后再基于身份子空间进行人脸识别,从而使年龄的干扰得到了有效的抑制.通过在MORPH和FGNET数据库上的实验,证实了基于年龄不变的身份特征子空间学习方法能提升人脸识别性能.     

人脸表情识别方法

介绍了表情识别特点及研究现状;重点阐述了不同的人脸表情特征提取方法和基于分类器的表情识别方法,并对各种方法进行了简单的分析比较;最后针对鲁棒性的需求,给出了人脸表情识别未来要研究的重点内容。     

人脸识别中的眼睛定位

本文提出一种结合人脸中眼睛的灰度及形状信息对人脸识别中的眼睛进行定位的算法。该算法容易实现,运算量少,且与图像的平移、旋转、尺度无关。实验表明,该算法在人脸识别的眼睛定位上有较高的准确度。     

IFRS: An Indexed Face Recognition System Based on Face Recognition and RFID Technologies

Access control systems are in contact with humans in everyday life, it is used in buildings, smartphones, cars, and IoT. Access control systems became an active research area. The performance of an access control system is specified by its speed and accuracy. Biometric systems are powerful access control systems which use humans’ biological or physiological properties to provide access to the restricted data or area. From all of the many biometric system types, the face recognition system is the only type that is delivering the automatic property. Moreover, it is the most acceptable type of biometric systems to the humans. The main challenges in the face recognition system are the degradation of the speed and accuracy when the system database grew bigger. This is because the face recognition system is an identification system that adopts a one to many (1:M) relationship. As a result, there is a need to develop a system with one to one (1:1) relationship, which is a challenging process. Motivated by such challenge, this paper proposes a system called Indexed Face Recognition System (IFRS) which is based on the combination of face recognition technology and Radio Frequency Identification technology. IFRS uses Local Binary Pattern Histogram as a feature vector and Haar-cascade classifier for the face detection. Moreover, the system is enhanced with three pre-processing methods namely: Bilateral filter, Histogram Equalization, and applying Tan and Triggs’ algorithm. In addition, IFRS performs an image normalization processes before and after Face Detection phase to enhance images quality, these process are: Color Conversion and Image Cropping and Resizing. Two experiments were done. The first experiment was done on 400 images with 40 subjects (10 images per subject). The second experiment was done on 210 collected images for 21 subjects (10 images per subject) from University students as a real-life case study. The practical results demonstrates that 4?×?4 image divisions gives better results than 8?×?8 image divisions as far as recognition time, database access time, and storage capacity are concerned. The practical results show that IFRS can reach an accuracy of 100% with a very little amount of time delay that is negligible.     

人脸识别方法研究

首先介绍了计算机人脸识别的4类主要方法，然后重点介绍了基于本征脸法的人脸识别技术的基本原理和实现步骤，最后通过实验比较了各类方法的识别率，总结出各种方法的优缺点，并对未来的识别技术进行了展望。     

基于仿生模式识别与传统模式识别的 人脸识别效果比较研究

本文实现了一种基于仿生模式识别的人脸识别系统,并将其识别效果同最近邻分类器与不同核函数的SVM进行了分析比较.以ORL人脸库为识别对象,针对有"拒识"的情况下,通过改变不同识别算法的可调参数,在保证参与训练人的正确识别率在大致相同水平的条件下,分析了参与训练人的错误识别率(错识别为参与训练的其他人)与未参与训练人的错误接受率(错识别为参与训练的某人)的优劣.比较结果表明,基于仿生模式识别的方法明显优于其它模式识别方法.     

人脸识别方法的综述

人脸识别技术是利用计算机分析人脸图像,从中提取有效的特征信息,用来辨认身份的一门技术。在此简单介绍了近年来常用的几种人脸识别方法,如：特征脸、隐马尔科夫、神经网络、以及弹性图匹配方法等。     

人脸识别研究进展

人脸识别技术在各种应用的推动下发展很快。而近几年中，为避免各种实际应用环境的限制，非约束环境下的人脸识别也得到了很大的发展。本文概述了人脸识别的一些主要方法，讨论了影响人脸识别的几个主要因素及其处理方法以及基于视频的人脸识别技术，最后指出了鲁棒的人脸识别技术的研究方向和发展前景。