文件名称:Face-Detector-Training
介绍说明--下载内容均来自于网络,请自行研究使用
由一个3D变形人脸模型取自动生成适应的训练样本。由统计视角,tailored训练数据保证了所有的数据变化且由任意的人脸属性丰富训练样本,例如,年龄或体重。更进一步,它可能自动适应到环境约束,例如,来自于监控摄像机的照明或视角约束。我们使用裁剪的(tailor)图象训练一个新的Viola Jones的adaboost 目标检测框架的多核实现。这个新的实现不仅快速的,而且多特征通道的使用成为可能,例如,在训练期间的颜色特征。在我们实验中,我们训练7个依赖视角的人脸检测子并在Face Detection Data Set 和 Benchmark (FDDB)中评估它们。- takes a look
into the automated generation of adaptive training samples
a 3D morphable face model. Using statistical insights,
the tailored training data guarantees full data variability
and is enriched by arbitrary facial attributes such as age
or body weight. Moreover, it can automatically adapt to
environmental constraints, such as illumination or viewing
angle of recorded video footage surveillance cameras.
We use the tailored imagery to train a new many-core implementation of Viola Jones’ AdaBoost object detection fr a mework. The new implementation is not only faster but also
enables the use of multiple feature channels such as color
features at training time.
into the automated generation of adaptive training samples
a 3D morphable face model. Using statistical insights,
the tailored training data guarantees full data variability
and is enriched by arbitrary facial attributes such as age
or body weight. Moreover, it can automatically adapt to
environmental constraints, such as illumination or viewing
angle of recorded video footage surveillance cameras.
We use the tailored imagery to train a new many-core implementation of Viola Jones’ AdaBoost object detection fr a mework. The new implementation is not only faster but also
enables the use of multiple feature channels such as color
features at training time.
(系统自动生成,下载前可以参看下载内容)
下载文件列表
Fast Face Detector Training Using Tailored Views-Scherbaum_2013_ICCV_paper.pdf