本文共 4870 字，大约阅读时间需要 16 分钟。

在Android OpenGLES2.0（十四）中，我们成功实现了Obj格式3D模型的加载。具体来说，我们加载了一个没有贴图和光照处理的帽子模型，通过手动添加光照来呈现立体效果。对于具有贴图和光照处理的模型，如何实现加载呢？让我们深入探讨一下。

模型及贴图加载

模型加载与之前的实现类似，但这次我们需要同时加载顶点法线和贴图坐标，并将它们传递给着色器。此外，还需要提取并解析mtl文件中的材质信息。解析流程如下：

技术实现细节

为了实现上述功能，我们设计了以下类：

public class MtlInfo {    public String newmtl;    public float[] Ka = new float[3]; // 阴影色    public float[] Kd = new float[3]; // 固有色    public float[] Ks = new float[3]; // 高光色    public float[] Ke = new float[3]; // Ke    public float Ns; // shininess    public String map_Kd; // 固有纹理贴图    public String map_Ks; // 高光纹理贴图    public String map_Ka; // 阴影纹理贴图    public int illum; // 光照模型}

public class Obj3D {    public FloatBuffer vert;    public int vertCount;    public FloatBuffer vertNorl;    public FloatBuffer vertTexture;    public MtlInfo mtl;    private ArrayList
   
     tempVert;    private ArrayList
    
      tempVertNorl;    private ArrayList
     
       tempVertTexture;    public int textureSMode;    public int textureTMode;    public void addVert(float d) {        if (tempVert == null) {            tempVert = new ArrayList<>();        }        tempVert.add(d);    }    public void addVertTexture(float d) {        if (tempVertTexture == null) {            tempVertTexture = new ArrayList<>();        }        tempVertTexture.add(d);    }    public void addVertNorl(float d) {        if (tempVertNorl == null) {            tempVertNorl = new ArrayList<>();        }        tempVertNorl.add(d);    }    public void dataLock() {        if (tempVert != null) {            setVert(tempVert);            tempVert.clear();            tempVert = null;        }        if (tempVertTexture != null) {            setVertTexture(tempVertTexture);            tempVertTexture.clear();            tempVertTexture = null;        }        if (tempVertNorl != null) {            setVertNorl(tempVertNorl);            tempVertNorl.clear();            tempVertNorl = null;        }    }    public void setVert(ArrayList
      
        data) {        int size = data.size();        ByteBuffer buffer = ByteBuffer.allocateDirect(size * 4);        buffer.order(ByteOrder.nativeOrder());        vert = buffer.asFloatBuffer();        for (int i = 0; i < size; i++) {            vert.put(data.get(i));        }        vert.position(0);        vertCount = size / 3;    }    public void setVertNorl(ArrayList
       
         data) {        int size = data.size();        ByteBuffer buffer = ByteBuffer.allocateDirect(size * 4);        buffer.order(ByteOrder.nativeOrder());        vertNorl = buffer.asFloatBuffer();        for (int i = 0; i < size; i++) {            vertNorl.put(data.get(i));        }        vertNorl.position(0);    }    public void setVertTexture(ArrayList
        
          data) { int size = data.size(); ByteBuffer buffer = ByteBuffer.allocateDirect(size * 4); buffer.order(ByteOrder.nativeOrder()); vertTexture = buffer.asFloatBuffer(); for (int i = 0; i < size; i++) { vertTexture.put(data.get(i)); } vertTexture.position(0); }}
        
       
      
     
    
   

光照与纹理处理

在顶点着色器中，我们需要传递以下参数到片元着色器：

attribute vec3 vPosition;attribute vec2 vCoord;uniform mat4 vMatrix;uniform vec3 vKa;uniform vec3 vKd;uniform vec3 vKs;varying vec2 textureCoordinate;attribute vec3 vNormal;varying vec4 vDiffuse;varying vec4 vAmbient;varying vec4 vSpecular;void main() {    gl_Position = vMatrix * vec4(vPosition, 1);    textureCoordinate = vCoord;    vec3 lightLocation = vec3(0.0, -200.0, -500.0);    vec3 camera = vec3(0, 200.0, 0);    float shininess = 10.0;    vec3 newNormal = normalize((vMatrix * vec4(vNormal + vPosition, 1)).xyz - (vMatrix * vec4(vPosition, 1)).xyz);    vec3 vp = normalize(lightLocation - (vMatrix * vec4(vPosition, 1)).xyz);    vDiffuse = vec4(vKd, 1.0) * max(0.0, dot(newNormal, vp));    vec3 eye = normalize(camera - (vMatrix * vec4(vPosition, 1)).xyz);    vec3 halfVector = normalize(vp + eye);    float nDotViewHalfVector = dot(newNormal, halfVector);    float powerFactor = max(0.0, pow(nDotViewHalfVector, shininess));    vSpecular = vec4(vKs, 1.0) * powerFactor;    vAmbient = vec4(vKa, 1.0);}

precision mediump float;varying vec2 textureCoordinate;uniform sampler2D vTexture;varying vec4 vDiffuse;varying vec4 vAmbient;varying vec4 vSpecular;void main() {    vec4 finalColor = texture2D(vTexture, textureCoordinate);    gl_FragColor = finalColor * vAmbient + finalColor * vSpecular + finalColor * vDiffuse;}

渲染实现

在应用程序中，我们可以按照以下步骤进行渲染：

List
   
     model = ObjReader.readMultiObj(this, "assets/3dres/pikachu.obj");List
    
      filters = new ArrayList<>();for (int i = 0; i < model.size(); i++) {    Obj3D obj = model.get(i);    obj.dataLock();    // 添加渲染相关逻辑}
    
   

总结

通过以上实现，我们成功地将带有贴图和光照处理的Obj格式3D模型加载到Android OpenGLES2.0环境中。整个过程包括Obj文件解析、mtl文件解析、顶点、法线和贴图数据的提取与处理，以及顶点着色器和片元着色器的实现。通过合理的数据传递和光照计算，我们能够实现高质量的3D渲染效果。

转载地址：http://yngs.baihongyu.com/