结合形变模型与图像修复的人脸姿态矫正

目的 人脸姿态偏转是影响人脸识别准确率的一个重要因素,本文利用3维人脸重建中常用的3维形变模型以及深度卷积神经网络,提出一种用于多姿态人脸识别的人脸姿态矫正算法,在一定程度上提高了大姿态下人脸识别的准确率。方法 对传统的3维形变模型拟合方法进行改进,利用人脸形状参数和表情参数对3维形变模型进行建模,针对面部不同区域的关键点赋予不同的权值,加权拟合3维形变模型,使得具有不同姿态和面部表情的人脸图像拟合效果更好。然后,对3维人脸模型进行姿态矫正并利用深度学习对人脸图像进行修复,修复不规则的人脸空洞区域,并使用最新的局部卷积技术同时在新的数据集上重新训练卷积神经网络,使得网络参数达到最优。结果 在LFW（labeled faces in the wild）人脸数据库和StirlingESRC（Economic Social Research Council）3维人脸数据库上,将本文算法与其他方法进行比较,实验结果表明,本文算法的人脸识别精度有一定程度的提高。在LFW数据库上,通过对具有任意姿态的人脸图像进行姿态矫正和修复后,本文方法达到了96.57%的人脸识别精确度。在StirlingESRC数据库上,本文方法在人脸姿态为±22°的情况下,人脸识别准确率分别提高5.195%和2.265%;在人脸姿态为±45°情况下,人脸识别准确率分别提高5.875%和11.095%;平均人脸识别率分别提高5.53%和7.13%。对比实验结果表明,本文提出的人脸姿态矫正算法有效提高了人脸识别的准确率。结论 本文提出的人脸姿态矫正算法,综合了3维形变模型和深度学习模型的优点,在各个人脸姿态角度下,均能使人脸识别准确率在一定程度上有所提高。     

基于多任务学习的多姿态人脸重建与识别

针对当前人脸识别中姿态变化会影响识别性能,以及姿态恢复过程中脸部局部细节信息容易丢失的问题,提出一种基于多任务学习的多姿态人脸重建与识别方法——多任务学习堆叠自编码器（MtLSAE）。该方法通过运用多任务学习机制,联合考虑人脸姿态恢复和脸部局部细节信息保留这两个相关的任务,在步进逐层恢复正面人脸姿态的同时,引入非负约束稀疏自编码器,使得非负约束稀疏自编码器能够学习到人脸部的部分特征;其次在姿态恢复和局部信息保留两个任务之间通过共享参数的方式来学习整个网络框架;最后将重建出来的正脸图像通过Fisherface进行降维并提取具有判别信息的特征,并用最近邻分类器来识别。实验结果表明,MtLSAE方法获得了较好的姿态重建质量,保留的局部纹理信息清晰,而且与局部Gabor二值模式（LGBP）、基于视角的主动外观模型（VAAM）以及堆叠步进自编码器（SPAE）等经典方法相比,识别率性能得以提升。     

基于Shearlet变换和均匀局部二值模式特征的协作表示人脸识别算法

为了获得人脸图像中更丰富的纹理特征以提高人脸识别率,提出了一种基于Shearlet变换和均匀局部二值模式(ULBP)算子提取特征(Shearlet_ULBP特征)的协作表示方法--Shearlet_ULBP CRC用于人脸识别。首先,人脸图像通过Shearlet变换分解,得到多尺度多方向的幅值域图谱,再经过简单的平均融合,获得融合后的幅值域图谱;然后,通过ULBP算子结合分块的方法获得该Shearlet变换后融合图像的直方图特征;最后,结合协作表示的方法来分类所提取到的特征。该方法可以提取到图像更丰富的边缘以及纹理信息,在ORL、Extended Yale B和AR人脸数据库上进行测试,在图像无遮挡的情况下识别率都达到了99%以上,在有遮挡情况下也都达到了91%以上的识别率。实验结果表明,所提方法不仅对于光照、姿态和表情变化具备较强的鲁棒性,同时能在一定程度上处理人脸图像中存在遮挡的情形。     

Non-manual cues in automatic sign language recognition

Present work deals with the incorporation of non-manual cues in automatic sign language recognition. More specifically, eye gaze, head pose, and facial expressions are discussed in relation to their grammatical and syntactic function and means of including them in the recognition phase are investigated. Computer vision issues related to extracting facial features, eye gaze, and head pose cues are presented and classification approaches for incorporating these non-manual cues into the overall Sign Language recognition architecture are introduced.     

基于表观的归一化坐标系分类视线估计方法

视线估计能够反映人的关注焦点,对理解人类的情感、兴趣等主观意识有重要作用。但目前用于视线估计的单目眼睛图像容易因头部姿态的变化而失真,导致视线估计的准确性下降。提出一种新型分类视线估计方法,利用三维人脸模型与单目相机的内在参数,通过人脸的眼睛与嘴巴中心的三维坐标形成头部姿态坐标系,从而合成相机坐标系与头部姿态坐标系,并建立归一化坐标系,实现相机坐标系的校正。复原并放大归一化得到的灰度眼部图像,建立基于表观的卷积神经网络模型分类方法以估计视线方向,并利用黄金分割法优化搜索,进一步降低误差。在MPIIGaze数据集上的实验结果表明,相比已公开的同类算法,该方法能降低约7.4%的平均角度误差。     

动态时间规整优化局部分块匹配的人脸识别

针对现实人脸识别中姿势、光照、表情变化及遮挡等严重影响识别性能的问题,提出了一种基于动态时间规整优化局部分块匹配的户外人脸识别算法。将人脸图像划分成若干大小相等且互不重叠的局部小块;借助于光栅扫描顺序将各个小块按照前额、眼睛、鼻子、嘴巴和下巴的顺序连接成一个单一序列;计算查询人脸与注册人脸之间图像到类的距离,利用动态时间规整的设计思想寻找查询序列与所有注册序列之间的最佳对齐方式。在三个公开人脸数据库LFW、AR及YouTube上的实验验证了该方法的有效性及可靠性,实验结果表明,相比其他几种较为先进的人脸识别方法,该方法取得了更高的识别率,此外,该方法无需任何训练过程,计算成本低。     

基于三维矫正和相似性学习的无约束人脸验证

针对无约束条件下的人脸图像样本少、面部姿态变化大、被遮挡以及背景复杂等问题,提出一种结合三维人脸矫正与相似性学习相结合的人脸验证算法（sub-SL）。首先,通过三维人脸矫正方法对人脸图像进行姿态矫正,将图像中的人脸矫正为标准正面脸;其次,裁剪该正面脸的脸部相关区域,去除复杂的图像背景;最后,利用基于个体子空间的相似性学习方法对图像对之间的相似度进行度量,完成人脸验证。实验采用了几个以LFW（Labeled Faces in the Wild）数据库为基础的经过预处理操作（例如人脸矫正、裁剪等）后建立起来的数据库。在基于局部三值模式（LTP）的特征提取方法并且训练图像对数为625的实验中,sub-SL算法的识别率比利用马氏距离进行度量学习的算法sub-ML以及结合了马氏距离与相似性学习的度量学习算法sub-SML分别高出了15.6%和8.4%。实验结果表明,sub-SL算法能够有效提高无约束条件下人脸识别的准确率。     

A robust incremental clustering-based facial feature tracking

Emerging significance of person-independent, emotion specific facial feature tracking has been actively tracked in the machine vision society for decades. Among distinct methods, the Constrained Local Model (CLM) has shown significant results in person-independent feature tracking. In this paper, we propose an automatic, efficient, and robust method for emotion specific facial feature detection and tracking from image sequences. A novel tracking system along with 17-point feature model on the frontal face region has also been proposed to facilitate the tracking of human basic facial expressions. The proposed feature tracking system keeps patch images and face shapes till certain number of key frames incorporating CLM-based tracker. After that, incremental patch and shape clustering algorithms is applied to build appearance model and structure model of similar patches and similar shapes respectively. The clusters in each model are built and updated incrementally and online, controlled by amount of facial muscle movement. The overall performance of the proposed Robust Incremental Clustering-based Facial Feature Tracking (RICFFT) is evaluated on the FGnet database and the Extended Cohn-Kanade (CK+) database. RICFFT demonstrates mean tracking accuracy of 97.45% and 96.64% for FGnet and CK+ database respectively. Also, RICFFT is more robust by minimizing average shape distortion error of 0.20% and 1.86% for FGnet and CK+ (apex frame) database, as compared with classic method CLM.     

基于DTCWT和LBP的低分辨率人脸识别

针对短时傅里叶变换频率分辨率较差的缺点,提出一种基于双树复小波变换(DTCWT)和局部二进制模式(LBP)直方图的低分辨率人脸识别方法。使用DTCWT获得人脸图像的多尺度多方向的频率幅度响应,采用LBP获取频率幅度响应的统计直方图,通过基于统计的一致性模式得到更加紧凑的统计分布特征。实验结果表明,该方法在低分辨率人脸上可以达到较高的识别准确率。     

Passive Three Dimensional Face Recognition Using Iso-Geodesic Contours and Procrustes Analysis

We introduce a new model for personal recognition based on the 3-D geometry of the face. The model is designed for application scenarios where the acquisition conditions constrain the facial position. The 3-D structure of a facial surface is compactly represented by sets of contours (facial contours) extracted around automatically pinpointed nose tip and inner eye corners. The metric used to decide whether a point on the face belongs to a facial contour is its geodesic distance from a given landmark. Iso-geodesic contours are inherently robust to head pose variations, including in-depth rotations of the face. Since these contours are extracted from rigid parts of the face, the resulting recognition algorithms are insensitive to changes in facial expressions. The facial contours are encoded using innovative pose invariant features, including Procrustean distances defined on pose-invariant curves. The extracted features are combined in a hierarchical manner to create three parallel face recognizers. Inspired by the effectiveness of region ensembles approaches, the three recognizers constructed around the nose tip and inner corners of the eyes are fused both at the feature-level and the match score-level to create a unified face recognition algorithm with boosted performance. The performances of the proposed algorithms are evaluated and compared with other algorithms from the literature on a large public database appropriate for the assumed constrained application scenario.     

姿态和表情变化下的三维人脸标志点定位

三维人脸标志点定位在人脸识别、人脸跟踪、人脸建模、表情分析等方面具有非常重要的作用. 然而, 在姿态和表情变化很大的条件下进行标志点定位, 这仍然是一个很具挑战性的课题. 本文提出一种对姿态和表情不敏感的三维人脸标志点定位方法, 利用HK曲率分析检测出候选标志点, 根据对面部形状的先验知识, 提出一种基于人脸几何结构的分类策略对候选标志点进一步细分, 通过把候选标志点与人脸标志点模型进行匹配, 实现标志点的精确定位. 首先在CASIA数据集对该方法的标志点定位精度进行测试, 然后在UND/FRGC v2.0数据集对该方法与其他方法进行比较. 实验结果表明该方法在姿态和表情变化很大的条件下具有高精度和高鲁棒性.     

一种复杂光照下的人脸识别算法

为提高光照变化下的人脸识别率,提出了一种基于局部对比增强（LCE）和局部相位量化（LPQ）的人脸识别方法。采用面部对称的思想结合LCE算法对受不均匀光照的人脸图像进行光照补偿;利用LPQ算子对增强后的图片进行标记,并用分块离散余弦变换（DCT）进行降维;分块计算LPQ直方图序列作为人脸图像的特征描述向量,送入最近邻分类器进行分类识别。通过Yale B和CAS_PEAL数据库上的实验,证实了所提方法的有效性。     

Non-lambertian reflectance modeling and shape recovery of faces using tensor splines

Modeling illumination effects and pose variations of a face is of fundamental importance in the field of facial image analysis. Most of the conventional techniques that simultaneously address both of these problems work with the Lambertian assumption and thus fall short of accurately capturing the complex intensity variation that the facial images exhibit or recovering their 3D shape in the presence of specularities and cast shadows. In this paper, we present a novel Tensor-Spline-based framework for facial image analysis. We show that, using this framework, the facial apparent BRDF field can be accurately estimated while seamlessly accounting for cast shadows and specularities. Further, using local neighborhood information, the same framework can be exploited to recover the 3D shape of the face (to handle pose variation). We quantitatively validate the accuracy of the Tensor Spline model using a more general model based on the mixture of single-lobed spherical functions. We demonstrate the effectiveness of our technique by presenting extensive experimental results for face relighting, 3D shape recovery, and face recognition using the Extended Yale B and CMU PIE benchmark data sets.     

Eye detection in a facial image under pose variation based on multi-scale iris shape feature

The accurate location of eyes in a facial image is important to many human facial recognition-related applications, and has attracted considerable research interest in computer vision. However, most prevalent methods are based on the frontal pose of the face, where applying them to non-frontal poses can yield erroneous results.In this paper, we propose an eye detection method that can locate the eyes in facial images captured at various head poses. Our proposed method consists of two stages: eye candidate detection and eye candidate verification. In eye candidate detection, eye candidates are obtained by using multi-scale iris shape features and integral image. The size of the iris in face images varies as the head pose changes, and the proposed multi-scale iris shape feature method can detect the eyes in such cases. Since it utilizes the integral image, its computational cost is relatively low. The extracted eye candidates are then verified in the eye candidate verification stage using a support vector machine (SVM) based on the feature-level fusion of a histogram of oriented gradients (HOG) and cell mean intensity features.We tested the performance of the proposed method using the Chinese Academy of Sciences' Pose, Expression, Accessories, and Lighting (CAS-PEAL) database and the Pointing'04 database. The results confirmed the superiority of our method over the conventional Haar-like detector and two hybrid eye detectors under relatively extreme head pose variations.     

基于压缩金字塔核稀疏表示的人脸识别

人脸识别中光照、伪装及姿态等变化一直是富有挑战性的问题,其中特征提取是很关键的一步。为提高人脸识别率,结合压缩感知和空间金字塔模型,本文提出了一种新的特征提取方法,首先用尺度不变特征变换算法提取图像特征,然后与随机生成的字典进行稀疏编码,再用金字塔模型分层提取不同尺度空间的特征,并用最大池融合特征,最后运用核稀疏表示分类。在Extended Yale B,AR 和CMU PIE人脸数据库上的实验结果表明,该方法对于人脸图像的光照、伪装及姿态等变化有较强的鲁棒性,而且该算法有较快的运行速度。     

使用人眼几何特征的视线追踪方法

目的 视线追踪是人机交互的辅助系统,针对传统的虹膜定位方法误判率高且耗时较长的问题,本文提出了一种基于人眼几何特征的视线追踪方法,以提高在2维环境下视线追踪的准确率。方法 首先通过人脸定位算法定位人脸位置,使用人脸特征点检测的特征点定位眼角点位置,通过眼角点计算出人眼的位置。直接使用虹膜中心定位算法的耗时较长,为了使虹膜中心定位的速度加快,先利用虹膜图片建立虹膜模板,然后利用虹膜模板检测出虹膜区域的位置,通过虹膜中心精定位算法定位虹膜中心的位置,最后提取出眼角点、虹膜中心点等信息,对点中包含的角度信息、距离信息进行提取,组合成眼动向量特征。使用神经网络模型进行分类,建立注视点映射关系,实现视线的追踪。通过图像的预处理对图像进行增强,之后提取到了相对的虹膜中心。提取到需要的特征点,建立相对稳定的几何特征代表眼动特征。结果 在普通的实验光照环境中,头部姿态固定的情况下,识别率最高达到98.9%,平均识别率达到95.74%。而当头部姿态在限制区域内发生变化时,仍能保持较高的识别率,平均识别率达到了90%以上。通过实验分析发现,在头部变化的限制区域内,本文方法具有良好的鲁棒性。结论 本文提出使用模板匹配与虹膜精定位相结合的方法来快速定位虹膜中心,利用神经网络来对视线落点进行映射,计算视线落点区域,实验证明本文方法具有较高的精度。     

Application notes - Algorithms for Assessing the Quality of Facial Images

In this paper, we presented algorithms to assess the quality of facial images affected by factors such as blurriness, lighting conditions, head pose variations, and facial expressions. We developed face recognition prediction functions for images affected by blurriness, lighting conditions, and head pose variations based upon the eigenface technique. We also developed a classifier for images affected by facial expressions to assess their quality for recognition by the eigenface technique. Our experiments using different facial image databases showed that our algorithms are capable of assessing the quality of facial images. These algorithms could be used in a module for facial image quality assessment in a face recognition system. In the future, we will integrate the different measures of image quality to produce a single measure that indicates the overall quality of a face image     

基于多种LBP特征集成学习的人脸识别*

单一的特征与分类器只能对限定条件下的人脸进行较好的识别,当在非限定条件下(如光照、背景等发生变化时)将出现人脸识别率较低问题,针对该问题,提出了一种基于多种局部二进制特征集成学习的人脸识别算法。首先,使用监督梯度下降法 (SDM)对人脸特征点定位,应用中心对称局部二进制(CSLBP)算子提取每个特征点邻域特征,将所有人脸特征点邻域特征合成为精细的纹理特征;同时运用分区LBP直方图算法提取人脸区域的微观空间结构特征;然后,使用K最近邻算法(KNN)和支持向量机(SVM)分别训练这两种特征,得到类别排序列表和投票决策矩阵;最后,利用加权求和的规则融合决策矩阵,构成最优集成分类器,从而得到输出类别。通过在非限制性人脸库LFW上实验结果表明,所提算法采用集成的方法明显优于单一的特征和分类器。     

低分辨率人脸图像的迭代标签传播识别算法

针对监控视频下低分辨率人脸识别中存在的特征表示能力不强及判别开集人脸图像身份不够准确的问题,文中提出低分辨率人脸图像的迭代标签传播识别算法.采用视觉几何组(VGG)提取人脸图像特征,依据特征的相似度获得高、低分辨率图像的映射关系.对已标记样本和未标记样本进行迭代标签传播,在每次迭代过程中,通过统计每个类别的置信度直方图,估计识别精确率逼近100%的自适应置信度阈值.根据阈值将确认的未标记样本更新至已标记样本集,提高算法召回率.在公共数据集上的实验表明,文中算法在精确率逼近100%的基础上,召回率取得较高值.