搜索资源列表
ComponentbasedFaceDetection
- Abstract We present a component-based, trainable system for detecting frontal and near-frontal views of faces in still gray images. The system consists of a two-level hierarchy of Support Vector Machine (SVM) classifiers. On the first level,
PersonIDFetch
- 研究描述人脸特征的有效方法, 讨论身份证照片的特征提取和检索采用自适应肤色检测技术改进通用的肤色检测算法, 进行脸部区域的划分提出系数投影法对面部五官区域进行分割, 在各区域中提取面部几何特征引人描述脸颊和下额轮廓的曲线参数作为脸形特征, 得到对人脸特征更准确的描述将面部几何特征矢量匹配、脸形曲线参数匹配和脸部图像相关匹配相结合, 实现人像照片的准确检索实验表明该方法性能优良。-Describe the facial features of an effective way to discuss
renlianshibieMATLAB
- 本文针对复杂背景下的彩色正面人脸图像,将肤色分割、模板匹配与候选人脸图像块筛选结合起来,构建了人脸检测实验系统,并用自制的人脸图像数据库在该系统下进行了一系列的实验统计。-Color frontal face images under complex background, skin color segmentation, template matching candidates face image block screening combined to build a face detect
FaceRecognitionBased-OnDeepLearning
- 本文运用深度神经网络的方法克服姿态变量和图像分辨率的影响,提出了一种多姿态的人脸超分辨识别算法并在实验数据集上获得了良好的性能表现。另外本文利用深度信念网络探索正面人脸图像和侧面人脸图像的映射,方法放松了深度信念网络的输入也输出之间绝对相等,而只是保证其高层含义上的相等。实验表明了使用深度信念网络可以学习到侧面人脸图像到正面人脸图像的一个全局映射,但是丢失了个体细节差异。本文还提出了基于深度网络保持姿态邻域进行姿态分类的方法,在学习过程中,我们保证了同一个姿态下的人脸图像应该与同一姿态下的多张图
AR
- AR人脸数据库,用于图像检索和图像分类,是BMP格式的,还包括MAT格式.-The AR database consists of over 4000 frontal images from 126 individuals. For each individual, 26 pictures were taken in tow separate sessions
multisvm
- Frontal views of all subjects are videotaped under constant illumination using fixed light sources, and none of the subjects wear eyeglasses. These constraints are imposed to minimize optical flow degradation. Previously untrained subjects ar
att_faces.tar
- Name: Olivetti - Att - ORL Color Images: No Image Size: 92 x 112 Number of unique people: 40 Number of pictures per person: 10 Different Conditions: All frontal and slight tilt of the head Citation reference: Ferdinando Samaria, Andy Hart