人脸图像的庞大信息量使其不适合于直接识别。
A great quantity of information in the face image makes it improper to be used to direct recognition.
人脸图像合成是新一代人机交互中的重要技术。
Facial expression synthesis is an important technique in human computer interactions.
本文着重介绍了基于学习的人脸图像超分辨率算法。
Then, we propose a new learning-based super-resolution algorithm for face images.
本文提出了一种解决滤除人脸图像中高频及低频噪声的新算法。
In this paper, we propose a new algorithm for removing high frequency and low frequency noise in face images.
提出了一种利用模板匹配与遗传算法的人脸图像特征提取的方法。
A method of extracting eye and mouth features from facial images using deformable templates and genetic algorithms is described.
对OR L人脸图像库的实验结果表明所给方法是可行的、有效的。
The results of experiments at ORL face image database show that the proposed method is feasible and effective.
针对表情正常的正面人脸图像，提出了一种基于图像差值的识别算法。
This paper proposes a kind of face recognition algorithm based on image difference among the front face pictures with normal expression.
人脸轮廓提取是人脸特征检测和人脸识别等人脸图像分析的重要前提。
Face contour extraction is a significant precondition of facial image analysis such as facial features detection and face recognition.
基于积分投影的人脸图像特征点的提取方法对人脸进行定位特别精确。
Bonus point projection based on the characteristics of face images from the point of methodology is precised for people face special position.
文中给出了用四通道结构，以正面人脸图像作为测试对象的实验结果。
A four channel structure is constructed and the experimental results in which frontal images of human faces are used as the subjects are given.
该系统对于视频序列和大容量的人脸数据库能够实时进行人脸图像检索。
To the given video sequence, the system can retrieve face image from big face image database on time.
实验结果表明，与其它算法比较，此算法能够对人脸图像进行良好的分类识别。
The experimental results show that, the algorithm can be run well on face image classification and recognition comparing with other algorithms.
为了构造自动处理人脸图像的信息系统，首先需要鲁棒、有效的人脸检测算法。
To build fully automated systems that analyze the information contained in face images, robust and efficient face detection algorithms are essentially required.
在接下来的测试中，科学家们给狗狗展示了一组它们在训练中从未见过的人脸图像。
In a subsequent test, the scientists showed the dogs images of human faces they had not seen in their training.
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。
According to features of edge gradient maps of face images, a new eye feature extraction algorithm based on gradient vector flow field was proposed.
六名志愿者被展示了人脸图像，这些图像展现不同的情感，诸如幸福、恐惧和惊讶。
Six volunteers were presented with images of people's faces, displaying different emotions such as happiness, fear and surprise.
实验结果表明，算法复原出的人脸图像更接近于真实图像，具有更高的峰值信噪比。
The experimental results demonstrate that the results of the proposed method are closer to the real images, with a higher peak signal to noise ratio.
在ORL标准人脸库及现实人脸图像上的实验结果表明该方法具有很好的鉴别分析能力。
The experimental results on ORL face database show that the method proposed has very good classification capability and high recognition rate.
最后对光学识别结果进行判决，从图像库中找出一个或几个与待识别目标相似的人脸图像。
The normalized image will then be encoded to an extensive complementary encoding image and be sent to an incoherent correlator based optical processor to be recognized.
对人脸图像检测与识别技术的研究现状进行简要论述，并自行设计一个人脸特征自动提取系统。
This paper examines the research situation on face image detection and recognition and designs a facial feature automatic extract system.
该算法首先计算人脸图像的相似度信息，并对得到的相似度图像进行二值化，从而标定人脸区域。
First the algorithm calculates the similarity information and gets the binary image to locate the face region.
分析了图像信号相位信息的空间特性，以人脸图像为例，研究了相位信息在图像特征定位中的应用。
The space characteristic of phase information of image signal is analysed, and using face image as an example, the application of phase information in image feature location is studied.
在侧面人脸图像检测过程中，利用积分投影法对预处理后的二值化侧面人脸图像进行人脸区域的定位。
During side face image detection, the person's face area is located by use of integration projection in the pre-disposed binary image.
介绍了一种新颖的人脸检测方法，该方法基于非统一背景下室内环境中单幅人脸图像的色度特征信息。
A novel human face detection method is introduced, which is based on chrominance colour information feature from an image containing one face in indoor environment with non-uniform background.
人脸识别中，人脸图像的信息不仅存在于像素间的二阶统计特性中，更存在于像素间的高阶统计特性中。
In the face recognition, important information not only exists in the picture element in two-order statistical property but also contains in the high-order statistical property.
对其中的人脸图像标准化方法、归一化处理和由单视图生成多姿态人脸图像等关键技术进行了重点探讨。
Besides, the paper focuses on the standardization of face image, the normalization and synthesis of multi-pose face image from a single view.
研究者给受试对象看大量的人脸图像，从很具有男性特征的到很具有女性特征的，然后让他们判断性别。
The researchers showed subjects a random series of faces, ranging along a spectrum of very male to very female, and asked them to classify the faces by gender.
基于所测量到的皮肤色泽的变化，研究人员建立了人脸图像，用于对其他学生的健康和外貌吸引力进行分级。
Based on the measured color changes, the researchers created face images that other students rated for health and attractiveness.
采取训练的方法得到面部动画参数和人脸图像之间的关系，直接进行人脸特征的跟踪和面部动画参数的抽取。
This paper trains the relationship of face animation parameters and different face images, and directly doing face animation parameters extraction and tracking.
采取训练的方法得到面部动画参数和人脸图像之间的关系，直接进行人脸特征的跟踪和面部动画参数的抽取。
This paper trains the relationship of face animation parameters and different face images, and directly doing face animation parameters extraction and tracking.
应用推荐