Question

在Qt论坛中，我找到了this question并激起了我的好奇心。

我在Qt中找到了一个非常简单的例子来显示一个立方体并修改了立方体逻辑以创建一个边长为1个单位的立方体。

然后我尝试点击模型并显示我点击的区域的坐标。

当没有涉及旋转时，似乎拾取工作正常。但！如果我取消评论paintGL中的轮换（或其中一个），我最终会得到“错误”的值。

例如：

没有旋转，几乎点击多维数据集最左边的边框：gluUnProject方法表示我点击了{ -0.49075, 0.234, 0.5 }点，这似乎没问题。

Cube without rotations

启用了旋转，并且几乎点击了多维数据集最左边的边框：我收到了点{ -0.501456, 0.157555, -0.482942 }，这似乎不对。 x坐标的范围不在-0.5, 0.5。

Cube with the two rotations enabled

我认为坐标转换是可以的。我一直在通过谷歌进行研究，人们总是使用相同的代码。此外，像素的颜色信息与我点击的脸部颜色相匹配。

那么，任何人都可以告诉我为什么在涉及旋转时我会得到错误的坐标？我认为我在基本的3D理解方面失败了，但我无法意识到它在哪里。

以下是代码：

main.cpp中：

#include <QApplication>
#include "GLCube.h"

int main(int argc, char *argv[])
{
    QApplication a(argc, argv);

    GLCube w;
    w.resize(800,600);
    w.show();

    return a.exec();
}

GLCube.h

#ifndef GLCUBE_H
#define GLCUBE_H

#include <QtOpenGL>
#include <QGLWidget>

class GLCube : public QGLWidget{
  Q_OBJECT // must include this if you use Qt signals/slots
public:
    GLCube(QWidget *parent = NULL)
        : QGLWidget(parent) {
    }
protected:
    // Set up the rendering context, define display lists etc.:
   void initializeGL();
   // draw the scene:
   void paintGL();
   // setup viewport, projection etc.:
   void resizeGL (int width, int height);
   virtual void mousePressEvent(QMouseEvent *pme);
};

#endif

GLCube.cpp：

#include "GLCube.h"

#include <cmath>
#include <iostream>
#include <iomanip>

#include "GL/glu.h"

namespace {
    float ver[8][3] =
    {
        { 0.5, -0.5, 0.5 },
        { -0.5, -0.5, 0.5 },
        { -0.5, -0.5, -0.5 },
        { 0.5, -0.5, -0.5 },
        { 0.5, 0.5, 0.5 },
        { -0.5, 0.5, 0.5 },
        { -0.5, 0.5, -0.5 },
        { +0.5, 0.5, -0.5 }
    };
    GLfloat color[8][4] =
    {
        {0.0,0.0,0.0, 1.0},
        {1.0,0.0,0.0, 1.0 },
        {1.0,1.0,0.0, 1.0 },
        {0.0,1.0,0.0, 1.0 },
        {0.0,0.0,1.0, 1.0 },
        {1.0,0.0,1.0, 1.0 },
        {1.0,1.0,1.0, 1.0 },
        {0.0,1.0,1.0, 1.0 },
    };

    void quad(int a,int b,int c,int d, int col)
    {
        glPointSize( 5 );
        glBegin(GL_POINTS);
            glColor4fv(color[1]);
            glVertex3fv(ver[a]);

            glColor4fv(color[2]);
            glVertex3fv(ver[b]);

            glColor4fv(color[3]);
            glVertex3fv(ver[c]);

            glColor4fv(color[4]);
            glVertex3fv(ver[d]);
        glEnd();

        glBegin(GL_LINES);
            glColor4fv(color[1]);
            glVertex3fv(ver[a]);
            glVertex3fv(ver[b]);

            glColor4fv(color[1]);
            glVertex3fv(ver[b]);
            glVertex3fv(ver[c]);

            glColor4fv(color[1]);
            glVertex3fv(ver[c]);
            glVertex3fv(ver[d]);

            glColor4fv(color[1]);
            glVertex3fv(ver[d]);
            glVertex3fv(ver[a]);
        glEnd();

        glBegin(GL_QUADS);
        glColor4fv(/*color[a]*/ color[col] );
        glVertex3fv(ver[a]);

    //    glColor3fv(color[b]);
        glVertex3fv(ver[b]);

    //    glColor3fv(color[c]);
        glVertex3fv(ver[c]);

    //    glColor3fv(color[d]);
        glVertex3fv(ver[d]);
        glEnd();
    }

    void colorcube()
    {
//        glPolygonMode(GL_FRONT_AND_BACK,GL_LINE);
        quad( 3, 2, 1, 0, 1 ); // bottom
        quad( 7, 6, 5, 4, 2); // top
        quad( 6, 2, 1, 5, 3 ); // front
        quad( 7, 3, 0, 4, 5 ); // back
        quad( 7, 6, 2, 3, 6 ); // left
        quad( 4, 5, 1, 0, 7); // right
    }
}
/*
 * Sets up the OpenGL rendering context, defines display lists, etc.
 * Gets called once before the first time resizeGL() or paintGL() is called.
 */
void GLCube::initializeGL(){
    //activate the depth buffer
    glEnable(GL_DEPTH_TEST);
    qglClearColor(Qt::black);
    glEnable(GL_CULL_FACE);
}


/*
 *  Sets up the OpenGL viewport, projection, etc. Gets called whenever the widget has been resized
 *  (and also when it is shown for the first time because all newly created widgets get a resize event automatically).
 */
void GLCube::resizeGL (int width, int height){
    glViewport( 0, 0, (GLint)width, (GLint)height );
    /* create viewing cone with near and far clipping planes */
    glMatrixMode(GL_PROJECTION);
    glLoadIdentity();
    glFrustum( -1.0, 1.0, -1.0, 1.0, 15.0, 30.0);

    glMatrixMode( GL_MODELVIEW );
}

/*
 * Renders the OpenGL scene. Gets called whenever the widget needs to be updated.
 */
void GLCube::paintGL(){

    //delete color and depth buffer
    glClear(GL_COLOR_BUFFER_BIT|GL_DEPTH_BUFFER_BIT);

    glMatrixMode(GL_MODELVIEW);
        glLoadIdentity();
        glTranslatef(0.0f,0.0f,-20.0f); //move along z-axis
        glRotatef(30.0,0.0,1.0,0.0); //rotate 30 degress around y-axis
        glRotatef(15.0,1.0,0.0,0.0); //rotate 15 degress around x-axis

    colorcube();
//    originalcube();
}

void GLCube::mousePressEvent(QMouseEvent *pme) {
    GLint viewport[4];
    GLdouble modelview[16];
    GLdouble projection[16];
    glGetDoublev( GL_MODELVIEW_MATRIX, modelview );
    glGetDoublev( GL_PROJECTION_MATRIX, projection );
    glGetIntegerv( GL_VIEWPORT, viewport );

    const int x = pme->x();
    const int y = viewport[3] - pme->y();

    qDebug() << "HERE: " << x << y;

    GLfloat color[4];
    glReadPixels( x, y, 1, 1, GL_RGBA, GL_FLOAT, color);
    GLenum error = glGetError();
    std::cout << "RETRIEVED COLOR:" << color[0] << ", " << color[1] << ", " << color[2] << ", " << color[3] << std::endl;
    printf( "\tERROR: %s (Code: %u)\n", gluErrorString(error), error );
    if(GL_NO_ERROR != error) throw;

    GLdouble depthScale;
    glGetDoublev( GL_DEPTH_SCALE, &depthScale );
    std::cout << "DEPTH SCALE: " << depthScale << std::endl;
    GLfloat z;
    glReadPixels( x, y, 1, 1, GL_DEPTH_COMPONENT, GL_FLOAT, &z );
    error = glGetError();
    std::cout << "X: " << x << ", Y: " << y  << ", RETRIEVED Z: " << z << std::endl;
    printf( "\tERROR: %s (Code: %u)\n", gluErrorString(error), error );
    if(GL_NO_ERROR != error) throw;
    std::cout << std::endl << std::endl;

    GLdouble posX, posY, posZ;
    GLint result;
    result = gluUnProject( x, y, z, modelview, projection, viewport, &posX, &posY, &posZ);
    error = glGetError();
    std::cout << "3D point with POS: " << posX << " " << posY << " " << posZ << std::endl;
    printf( "\tERROR: %s (Code: %u)\n", gluErrorString(error), error );
    std::cout << "\tglUnProject: " << (( GL_FALSE == result ) ? "FALSE" : "TRUE") << std::endl;
    if(GL_NO_ERROR != error) throw;
}

注意：

我在Windows 8.1 64位，Qt 4.8和MinGW中运行它。

此外，glReadPixels和gluUnProject都会毫无错误地退出（错误代码= GL_NO_ERROR）

并且glut不可用。仅限OpenGL，QtOpenGL和/或glu提供的内容。

Answer 1

这不是真的错，您正在读取深度缓冲区以找出用于反向投影的窗口空间Z值。

问题在于深度缓冲区中可用的精度有限，并且引入了一些不准确性。实际上，您不能指望未投影值的范围完全[ -0.5 ， 0.5 ]。你将不得不在这里引入一个小epsilon，所以你的有效范围将是[ -0.5015 ， 0.5015 ]。

您可以通过提高深度缓冲区的精度和/或减小近和远剪裁平面之间的范围来减轻影响。深度缓冲区默认情况下通常为24位定点，但32位定点或浮点深度缓冲区可能会略微改善您的情况。但是，您永远不会完全消除此问题。

当涉及旋转时，将鼠标坐标转换为OpenGL中的模型坐标

1 个答案: