关于透视投影的一些困惑

时间:2012-07-30 06:58:23

标签: opengl graphics perspective

我刚刚了解了Perspective投影,并且发现在openGl中应用它有点令人困惑。

考虑一个简单的广场 在使用透视投影之前,我可以在[-1,1]空间中定义其顶点的坐标,并输入{0.0f,0.0f,0.0f,1.0f,1.0f,1.0f,1.0f,0.0 f},方块将占据窗口的第一象限 请考虑我的代码中的以下部分:

设置矩阵:

    glm::mat4 mvp(1.0);
    mvp*=glm::perspective(45.0f,1.0f,0.01f,100.0f);

制服:

    loc = glGetUniformLocation(program.getHandle(),"mvp");
    glUniformMatrix4fv(loc,1,GL_FALSE,glm::value_ptr(mvp));

顶点着色器:

    #version 330
    in vec2 pos;
    uniform mat4 mvp;
    void main()
    {
        gl_Position = mvp*vec4(pos,0.0f,1.0f);
    }

但结果是一个空白的屏幕 我需要在这里应用哪些额外的转换?
另外,顶点数据应该在哪个坐标中?

以下是完整的代码:

#include "program.h"
#include <GL/glew.h>
#include <SFML/Window.hpp>
#include <stdexcept>
#include <iostream>
#include <glm/glm.hpp>
#include <glm/gtc/type_ptr.hpp>
#include <glm/gtc/matrix_transform.hpp>
#include <glm/core/type.hpp>

void GlewInit()
{
        GLenum err = glewInit();
        if (GLEW_OK != err)
        {
                std::cerr<<"Error: "<<glewGetErrorString(err)<<std::endl;
                throw(std::runtime_error("GLEW_INIT Failure."));
        }
}
int main()
{
    sf::Window win(sf::VideoMode(400,400),"Manasij");
    GlewInit();
    float vertexdata[]={0.0f,0.0f,0.0f,1.0f,1.0f,1.0f,1.0f,0.0f};
    GLubyte indexdata[]={0,1,2,3};
    float colordata[]={0.3f,0.0f,0.5f,1.0f};
    mm::Program program
    (
        {
            mm::Shader(GL_VERTEX_SHADER,"vshader.vert"),
            mm::Shader(GL_FRAGMENT_SHADER,"fshader.frag")
        }
    );

    glm::mat4 mvp(1.0);

    mvp*=glm::perspective(45.0f,1.0f,0.01f,100.0f);


    GLuint vao,vbo,ibo;
    glGenVertexArrays(1,&vao);
    glBindVertexArray(vao);

    glGenBuffers(1,&vbo);
    glBindBuffer(GL_ARRAY_BUFFER,vbo);
    glBufferData(GL_ARRAY_BUFFER,8*sizeof(GLfloat),vertexdata,GL_STATIC_DRAW);

    glGenBuffers(1,&ibo);
    glBindBuffer(GL_ELEMENT_ARRAY_BUFFER,ibo);
    glBufferData(GL_ELEMENT_ARRAY_BUFFER,4*sizeof(GLubyte),indexdata,GL_STATIC_DRAW);

    glVertexAttribPointer(0,2,GL_FLOAT,GL_FALSE,0,0);
    glEnableVertexAttribArray(0);
    glBindAttribLocation(program.getHandle(),0,"pos");


    glUseProgram(program.getHandle());
    auto loc = glGetUniformLocation(program.getHandle(),"col");
    glUniform4fv(loc,1,colordata);
    loc = glGetUniformLocation(program.getHandle(),"mvp");
    glUniformMatrix4fv(loc,1,GL_FALSE,glm::value_ptr(mvp));
    glUseProgram(0);


    glClearColor(1.0f,1.0f,1.0f,1.0f);
    glEnable(GL_DEPTH_TEST);
    glDepthFunc(GL_LESS);
    while(win.isOpen())
    {
        glClear(GL_COLOR_BUFFER_BIT);
        glUseProgram(program.getHandle());
//      glDrawArrays(GL_QUADS,0,4);
        glDrawElements(GL_QUADS,4,GL_UNSIGNED_BYTE,nullptr);
        glUseProgram(0);

        win.display();
        sf::Event eve;
        while(win.pollEvent(eve))
            if(eve.type==sf::Event::Closed)
                win.close();
    }
    return 0;
}

2 个答案:

答案 0 :(得分:4)

  

但是,结果是一个黑屏。   我需要在这里应用哪些额外的转换?

您会看到glm::perspective调用的最后两个参数,它们定义了近距离和远距离剪辑距离。你的方格在Z = 0,没有进一步的变换因此落在观看体积之外。解决方案:在-Z方向上移动seqare几个单位。

请注意near和far值的以下约束:

  • abs(远)&gt; abs(近)&gt; 0

  • sgn(near)= sgn(far)

如果 sgn(near)= sgn(far)= - ,您还必须交换深度范围的符号,或将深度测试功能更改为GL_LESS。

由于深度缓冲分辨率不是线性分布,而是以类似1 / z的方式分布

  • abs(near)应该尽可能大
  • abs(far)应该尽可能小。

答案 1 :(得分:0)

也许这个网站可以帮到你,它解释了OpenGL中投影矩阵的设置:

http://www.songho.ca/opengl/gl_projectionmatrix.html