一种基于遮挡区域补偿的眼镜去除方法

为了提高戴眼镜人脸图像的识别率, 提出了一种从人脸图像中检测并去除眼镜的方法。首先对输入的戴眼镜人脸图像与系统预留的无眼镜人脸图像进行基于人眼位置的标定, 检测出眼镜遮挡区域, 再用无眼镜人脸图像中对应的遮挡区域对戴眼镜人脸图像进行补偿, 从而合成了对应输入图像的不戴眼镜的人脸图像。实验结果表明, 该方法能有效地合成无眼镜人脸图像, 将合成后的人脸图像再应用于人脸识别系统, 识别率显著提高。     

基于可视信息眼镜摘除的热红外人脸识别方法

在热红外人脸识别中,眼镜作为人脸图像中常见的遮挡物,造成了人脸眼睛区域信息的丢失,严重影响了人脸识别效果。针对该问题,提出了一种在热红外图像中去除眼镜的算法,对热红外图像进行眼镜检测,使用无眼镜的热红外图像的平均眼睛模板来代替有眼镜的热红外图像的眼镜区域,再基于核主成分分析算法利用可视化图像和热红外图像融合的方法,进行图像融合,获得较好的无眼镜热红外图像,通过分类识别来实现人脸识别。实验结果表明,在热红外人脸识别中,该方法在戴眼镜的情况下能够提高人脸识别的准确率和取得较好的识别效果。     

人脸遮挡区域检测与重建

提出一种基于模糊主分量分析技术(FPCA)的人脸遮挡检测与去除方法.首先,有遮挡人脸被投影到特征脸空间并通过特征脸的线性组合得到一个重建人脸.计算重建图与原图的差图像,加权滤波后并归一化作为被遮挡的概率,以此概率为权重由原图和重建图合成新的人脸.在后续迭代中,根据遮挡概率使用模糊主分量分析进行分析重建,并使用累积误差进行遮挡检测.实验结果表明,算法可精确定位人脸遮挡区域,得到平滑自然的重建人脸图像,优于经典的迭代PCA方法.     

去除人脸图像中眼镜的方法

人脸图像的校正是人脸识别相关技术的前提性工作之一.眼镜作为一种常见的干扰物对识别结果有着重要的影响.针对该问题,提出了一种去除眼镜的面部图像校正方法.首先对无眼镜的一组人脸样本集进行训练,然后通过主成分分析法对输入的带眼镜的人脸图像进行重构,再按照原始图像的灰度特征对重构出的图像进行处理,最后合成为不带眼镜的人脸图像.实验结果表明,该算法易于实现,合成效果比较自然,有利于后续的处理过程.     

眼镜遮挡下的正面人脸识别

针对眼镜遮档对人脸识别影响较大这一问题,提出一种从正面人脸图像中提取并摘除眼镜的方法。首先利用主成分分析和独立成分分析法对输入的戴眼镜人脸进行重建,对比重建人脸和输入人脸,从而提取眼镜遮档区域;然后经过迭代误差补偿合成相应的无眼镜人脸;最后考虑到合成图像的特殊性,使用改进的特征加权方法实现人脸识别。实验结果表明,利用提出的人脸重建和特征加权方法进行戴眼镜人脸识别,正确率可以达到91%,优于传统方法。     

基于三维建模的眼镜遮挡下人脸识别

眼镜作为人脸特征的不稳定性是眼镜遮挡人脸识别的主要问题。为避免现有方法消除不稳定眼镜特征时带来的人脸特征丢失,将眼镜视为人脸固有部分,提出一种基于三维建模生成人脸虚拟样本补偿眼镜不稳定性的方法。三维建模方便眼镜模型参数的调节。通过调节眼镜参数,具体分析眼镜不同部分对人脸识别的影响,同时,针对影响严重的镜片模糊和反光,均做相应处理。CAL-PEAL的实验表明本文方法对识别性能的大幅度改善,并验证镜片处理的有效性。     

一种处理部分遮挡表情图像的方法

针对目前表情识别中眼部因头发、帽子等物体而存在部分遮挡的问题,提出了一种基于对称变换的眼部遮挡处理方法。方法针对二值化的人脸表情图像,参照人脸几何特征对眼部区域进行垂直积分投影;通过中心线检测算法,确定人脸的中心线并判断是否存在遮挡。对于不可容忍的遮挡,进行对称变换处理以修复表情图像。实验表明,在相同的特征提取方法和分类器选择情况下,该方法可有效提高部分遮挡人脸表情的识别效果,并可容忍头部一定范围内的偏转。     

基于掩膜生成网络的遮挡人脸检测方法

针对复杂遮挡条件下人脸检测精度低的问题,提出一种基于掩膜生成网络（MGN）的遮挡人脸检测方法。对人脸训练集进行预处理,将训练人脸划分为25个子区域,并为每个子区域分别添加遮挡。将一系列添加遮挡的人脸图像和原始人脸图像作为图像对依次输入MGN进行训练,以生成对应各个遮挡子区域的遮挡掩膜字典。通过组合相关字典项生成与检测人脸遮挡区域相对应的组合特征掩膜,并将该组合特征掩膜与检测人脸深层特征图相点乘,以屏蔽由局部遮挡引起的人脸特征元素损坏。在AR和MAFA数据集上进行实验,结果表明,该方法的检测精度高于MaskNet、RPSM等方法,且检测速度较快。     

基于深度对齐网络的生成对抗网络伪造人脸检测

针对现有的生成对抗网络（GAN）伪造人脸图像检测方法在有角度及遮挡情况下存在的真实人脸误判问题,提出了一种基于深度对齐网络（DAN）的GAN伪造人脸图像检测方法。首先,基于DAN设计面部关键点提取网络,以提取真伪人脸关键点位置;然后,采用主成分分析（PCA）方法将每一组关键点映射到三维空间,从而减少冗余信息以及降低特征维度;最后,利用支持向量机（SVM）五折交叉验证对特征进行分类,并计算准确率。实验结果表明,该方法通过提高面部关键点定位准确度改善了由于定位误差引起的面部不协调问题,进而降低了真实人脸误判率。与VGG19、XceptionNet和Dlib-SVM方法相比,正脸情况下,该方法的ROC下面积（AUC）值提高了4.48到32.96个百分点,平均精度（AP）提高了4.26到33.12个百分点;有角度及遮挡人脸情况下,该方法的AUC值提高了10.56到30.75个百分点,AP提高了7.42到42.45个百分点。     

一种快速眼睛定位与人脸平面旋转校正方法

针对人脸图像平面旋转导致识别效果不佳的问题,提出一种有效的眼睛定位与人脸平面旋转校正方法.首先基于AdaBoost算法训练得到的眼睛分类器从人脸图像中快速确定双眼候选区域,然后根据双眼在候选区中所处位置和所占比例,将候选区域分为宽度相同的左眼子区、中间子区和右眼子区3个子区域,再对包含眼睛的左眼子区和右眼子区分别求积分投影来准确定位双眼位置;准确定位双眼位置后,给出了以图像中心为旋转基准点时旋转角度的计算方法,并结合图像旋转公式实现了人脸图像的平面旋转校正.实验结果表明,该方法不仅能防止人脸倾斜时出现伪特征点,也能避免眼镜内边框和眼镜支架对定位眼睛的影响,提高眼睛定位精度,而且能实现人脸的平面旋转校正,具有非常好的实时性和实用价值.     

眼镜零售业计算机虚拟试戴方法研究

眼镜行业信息化建设主要涉及的两个方向是眼镜店销售管理系统和基于虚拟试戴技术的销售系统.计算机虚拟试戴方法研究是眼镜行业信息化研究的重点.该研究主要运用了计算机图形图像处理学的知识.实现了人脸与镜架按照真实比例的试戴,以及染色镜片效果的演示,是一种适合制作眼镜行业销售系统的方法.     

Human-centric design personalization of 3D glasses frame in markerless augmented reality

This paper presents a virtual try-on system based on augmented reality for design personalization of facial accessory products. The system offers several novel functions that support real-time evaluation and modification of eyeglasses frame. 3D glasses model is embedded within video stream of the person who is wearing the glasses. Machine learning algorithms are developed for instantaneous tracking of facial features without use of markers. The tracking result enables continuously positioning of the glasses model on the user’s face while it is moving during the try-on process. In addition to color and texture, the user can instantly modify the glasses shape through simple semantic parameters. These functions not only facilitate evaluating products highly interactive with human users, but also engage them in the design process. This work has thus implemented the concept of human-centric design personalization.     

Automatic eyeglasses removal from face images

In this paper, we present an intelligent image editing and face synthesis system that automatically removes eyeglasses from an input frontal face image. Although conventional image editing tools can be used to remove eyeglasses by pixel-level editing, filling in the deleted eyeglasses region with the right content is a difficult problem. Our approach works at the object level where the eyeglasses are automatically located, removed as one piece, and the void region filled. Our system consists of three parts: eyeglasses detection, eyeglasses localization, and eyeglasses removal. First, an eye region detector, trained offline, is used to approximately locate the region of eyes, thus the region of eyeglasses. A Markov-chain Monte Carlo method is then used to accurately locate key points on the eyeglasses frame by searching for the global optimum of the posterior. Subsequently, a novel sample-based approach is used to synthesize the face image without the eyeglasses. Specifically, we adopt a statistical analysis and synthesis approach to learn the mapping between pairs of face images with and without eyeglasses from a database. Extensive experiments demonstrate that our system effectively removes eyeglasses.     

Near infrared face recognition by combining Zernike moments and undecimated discrete wavelet transform

This study proposes a novel near infrared face recognition algorithm based on a combination of both local and global features. In this method local features are extracted from partitioned images by means of undecimated discrete wavelet transform (UDWT) and global features are extracted from the whole face image by means of Zernike moments (ZMs). Spectral regression discriminant analysis (SRDA) is then used to reduce the dimension of features. In order to make full use of global and local features and further improve the performance, a decision fusion technique is employed by using weighted sum rule. Experiments conducted on CASIA NIR database and PolyU-NIRFD database indicate that the proposed method has superior overall performance compared to some other methods in the presence of facial expressions, eyeglasses, head rotation, image noise and misalignments. Moreover its computational time is acceptable for on-line face recognition systems.     

Illumination invariant face recognition using near-infrared images

Most current face recognition systems are designed for indoor, cooperative-user applications. However, even in thus-constrained applications, most existing systems, academic and commercial, are compromised in accuracy by changes in environmental illumination. In this paper, we present a novel solution for illumination invariant face recognition for indoor, cooperative-user applications. First, we present an active near infrared (NIR) imaging system that is able to produce face images of good condition regardless of visible lights in the environment. Second, we show that the resulting face images encode intrinsic information of the face, subject only to a monotonic transform in the gray tone; based on this, we use local binary pattern (LBP) features to compensate for the monotonic transform, thus deriving an illumination invariant face representation. Then, we present methods for face recognition using NIR images; statistical learning algorithms are used to extract most discriminative features from a large pool of invariant LBP features and construct a highly accurate face matching engine. Finally, we present a system that is able to achieve accurate and fast face recognition in practice, in which a method is provided to deal with specular reflections of active NIR lights on eyeglasses, a critical issue in active NIR image-based face recognition. Extensive, comparative results are provided to evaluate the imaging hardware, the face and eye detection algorithms, and the face recognition algorithms and systems, with respect to various factors, including illumination, eyeglasses, time lapse, and ethnic groups     

基于差分投影和梯度向量流场的眼睛特征提取

提出了差分投影的方法确定眼睛区域,并根据人脸图像的边缘梯度图提出了一种新的基于梯度向量流场的眼睛特征提取方法。该算法改进原有的梯度向量流迭代方程求解梯度向量流场,以梯度向量流场中的汇点作为候选点,通过人脸器官的几何位置关系等方法评价候选点以定位眼球。算法能较好地容忍一定的光照变化、人脸的小角度倾斜和旋转、闭眼和眼镜等干扰。在具有以上干扰的ORL人脸库(400幅图像)和JAFFE人脸表情库(213幅图像)上的实验证明该算法具有较好的眼睛特征抽取能力。