如何规范从Kinect C＃中提取的功能？

Question

我正在做一个项目，用C＃中的Kinect提取嘴唇特征。我的问题是如何对功能进行归一化？

我认为我必须执行以下功能：

Nd =（Cd * Fd）/ dFN

其中Nd是归一化的距离，Cd是到Kinect的当前距离，Fd是每帧均被归一化的前额深度值，dFN是每帧都被归一化的深度值。

此外，还必须实现z分数归一化：

z =（X-μ）/σ

其中z是z得分归一化的结果，X是当前特征值，μ是在一个数据库中获得的平均值，σ表示方差。

如何根据功能进行归一化？

以下是我要归一化的功能：

private void FaceReader_FrameArrived(object sender, HighDefinitionFaceFrameArrivedEventArgs e)
{
    NewRecording();

    using (var frame = e.FrameReference.AcquireFrame())
    {
        if (frame != null)
        {
            frame.GetAndRefreshFaceAlignmentResult(_faceAlignment);
            UpdateFacePoints();

            var vertices = _faceModel.CalculateVerticesForAlignment(_faceAlignment);

            var lipLeftCorner = vertices[(int)HighDetailFacePoints.MouthLeftcorner];
            var lipRightCorner = vertices[(int)HighDetailFacePoints.MouthRightcorner];
            var lipWidth = Math.Abs(lipRightCorner.X - lipLeftCorner.X);
            var lipTopCorner = vertices[(int)HighDetailFacePoints.MouthUpperlipMidtop];
            var lipBottomCorner = vertices[(int)HighDetailFacePoints.MouthLowerlipMidbottom];
            var lipHeight = Math.Abs(lipTopCorner.Y - lipBottomCorner.Y);
            var chinCenter = vertices[(int)HighDetailFacePoints.ChinCenter];

            Console.WriteLine(lipLeftCorner.X.ToString());
            Console.WriteLine(lipRightCorner.X.ToString());
            Console.WriteLine(lipWidth.ToString());
            Console.WriteLine(lipTopCorner.Y.ToString());
            Console.WriteLine(lipBottomCorner.Y.ToString());
            Console.WriteLine(lipHeight.ToString());
            Console.WriteLine(lipTopCorner.Z.ToString());
            Console.WriteLine(chinCenter.X.ToString());
            Console.WriteLine(chinCenter.Y.ToString());
        }
    }
}