How to apply textures to a game character using OpenGL ES
Introduction
Now that you know how to render a character using OpenGL ES, you may want to learn how to apply images (textures) to a character. The process of applying a texture to a character is quite simple. We simply load texture coordinates into OpenGL buffers, along with the raw image representing the texture.
Objective
Our objective is to learn how to apply a texture to a character as seen in the image below.
This is a continuation tutorial from our previous tutorial, How to render a character in iOS devices.
This tutorial is a hands-on project. Feel free to download this template XCode project and code along.
Things to know
If this is your first time reading about textures in computer graphics, I recommend you read:
What is a Texture?
A texture is simply an image that clothes a character. Figure 1 shows a 3D model without texture and with texture. You can see that a texture adds personality to the model.
Figure 1. A model with and without texture.
What is a UV Map?
Let's assume that you, aside from being a game developer, are also a 3D modeler. You decide to model a nice looking robot on Blender, a 3D modeling software. Your intentions are to apply a texture to the robot, therefore, you unwrap the model to its 2D equivalent, as shown below.
Figure 2. Unwrapping a 3D model
By unwrapping the 3D character into a 2D entity, an image can properly be glued to the character. The process of unwrapping a 3D model into its 2D equivalent produces what is known as a UV Map. Figure 2 shows an example of a UV Map.
The process does not stop there. The UV Map is usually exported to an application like Photoshop or GIMP, where it is used as a blueprint to paint over a blank canvas. The resulting artwork is then saved as a .png file and imported back into Blender where you could see the final result. The resulting .png file, i.e., an image, is known as a Texture.
Figure 3. UV Map with image
What are UV Coordinates?
During the unwrapping process, the vertices of the 3D model are mapped into a two-dimensional coordinate system. This new coordinate system is known as the UV Coordinate System.
The UV Coordinate System is composed of two axes, known as U and V axes. These axes are equivalent to the X-Y axes you are familiar with. The only difference is that the U-V axes ranges from 0 to 1.
The new vertices produced by the unwrapping of the model are called UV Coordinates. These coordinates will be loaded into the GPU. And will serve as reference points to the GPU as it attaches the texture to the model.
Loading UV Coordinates
The loading of UV Coordinates into an OpenGL buffer is exactly similar as the loading of vertex data. We will use the same loading method employed in our previous tutorial: How to render a character in iOS.
Increase the size of the buffer
Since we are going to add UV coordinate data into the buffer, we need to allocate additional space to the buffer. The additional space required depends on the size of our _robotuv array.
The > > > robot> > > uv_> > array is found in the Robot.h file.
Open up file Character.mm. Locate the setupOpenGL() method. Go to line 5 and modify glBufferData() from what is shown in listing 1:
Listing 1 Previous glBufferData.
void Character::setupOpenGL(){
//...Binded VAO and Buffers
//5. Dump data into the buffer
glBufferData(GL_ARRAY_BUFFER, sizeof(robot_vertices)+sizeof(robot_normal), NULL, GL_STATIC_DRAW);
//...loaded data using glBufferSubData
}
to what is shown in listing 2.
Listing 2 Modified glBufferData.
void Character::setupOpenGL(){
//...Binded VAO and Buffers
//5. Dump data into the buffer
glBufferData(GL_ARRAY_BUFFER, sizeof(robot_vertices)+sizeof(robot_normal)+sizeof(robot_uv), NULL, GL_STATIC_DRAW);
//...loaded data using glBufferSubData
}
Load UV Coordinates
Now that we have allocated enough memory in the buffer, we can load the UV coordinates data. In the setupOpenGL method, go to line 5c and load the UV data as shown in listing 3.
Listing 3.
//5c. Load UV coordinates with glBufferSubData
glBufferSubData(GL_ARRAY_BUFFER, sizeof(robot_vertices)+sizeof(robot_normal), sizeof(robot_uv), robot_uv);
Get Location of the texture attribute
Next, we need to find the location of the attribute texCoord in the shader. We will use this location to link the UV coordinate data in the buffer to the texCoordattribute.
Go to line 8 and type what is shown in listing 4.
Listing 4.
//8. Get the location of the shader attribute called "texCoords"
uvLocation=glGetAttribLocation(programObject, "texCoord");
Enable the UV attribute
Next, we need to enable the UVattributes. That is, make the attribute available to accept data coming from the buffer.
Go to line 9c and type what is shown in listing 5.
Listing 5.
//9c. Enable the UV attribute
glEnableVertexAttribArray(uvLocation);
Link UV data to attribute location
Next, we are going to link the UV coordinate data in the buffer to the texCoord attribute location.
Go to line 10c and type what is shown in listing 6.
Listing 6.
//10c. Link the buffer data to the shader's UV location
glVertexAttribPointer(uvLocation, 2, GL_FLOAT, GL_FALSE, 0, (const GLvoid*)(sizeof(robot_vertices)+sizeof(robot_normal)));
Creating Texture Objects
Before we create an OpenGL Object designed to deal with textures, OpenGL requires the activation of a Texture-Unit. A texture-unit is the section in the GPU responsible for Texturing Operations.
After activating a texture-unit, any subsequent operation affects that particular texture-unit. Thus, subsequent creation of an OpenGL Object, makes the texture-unit the owner of the object.
Our first task is to activate a texture-unit, then create and bind an OpenGL Texture Object.
OpenGL objects designed to work with textures are called Texture Objects. Like any other OpenGL Object, Texture Objects require its behavior to be specified.
Texture Objects can either behave as objects that carry two-dimensional images or Cube Maps images. Cube Maps images are a collection of six 2D-images that form a cube; they are used to provide a sky scenery in a game.
Let's create a texture unit and a Texture Object. Locate the method setupOpenGL() and copy what is shown in listing 7.
A texture-unit is activated as shown in line 14. The generation and binding of a texture object as shown in line 15 & 16. In this example, we are defining the behavior of the texture object as a two-dimensional image.
Listing 7. Creating a Texture OpenGL Object
//14. Activate GL_TEXTURE0
glActiveTexture(GL_TEXTURE0);
//15.a Generate a texture OpenGL object
glGenTextures(1, &textureID[0]);
//16 Bind the texture OpenGL object. Any subsequent OpenGL operations, will apply to this object.
glBindTexture(GL_TEXTURE_2D, textureID[0]);
Loading the image into a Texture Object
Decompressing the image
Texture images are normally exported in .png, .tga or .jpeg formats. Unfortunately, these image-compression formats are not accepted by OpenGL. Images must be decompressed to a raw-format before they can be loaded into a texture object. The OpenGL API does not provide any decompression utility.
Fortunatelly, there exist various libraries which can be integrated in your application. One of them is called Lodepng. This library accepts a .png image and provides a decompressed version of the image.
In the setupOpenGL() method, type what is shown in listing 8. The method convertImageToRawImage() is a helper method, which calls a decompression method in the lodepng library. In the example below, we provide the name of our texture, in this case, red.png (line 17).
Listing 8. Converting images to raw data
//17. Decode image into its raw image data
if(convertImageToRawImage("red.png")){
//if decompression was successful, set the texture parameters
}
Setting Texture Parameters
If the decompression of the image is successful, Texture Parameters can be set. Texture Parameters inform the GPU how to apply the texture to the 3D model. That is, should it magnify or minimize the texture? We are going to ask OpenGL to clamp our image to the edges of our model and to do a linear interpolation between the pixels and texels.
In method setupOpenGL(). Type what is shown in lines 17a & 17b in listing 9.
Listing 9. Setting the texture parameters.
//17a. set the texture wrapping parameters
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
//17b. set the texture magnification/minification parameters
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER,GL_LINEAR);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
Texture Filtering
The geometry of a model usually does not have a one-to-one correspondence with the texture. The texture will need to be streched or shrunk to make it fit the model. To provide a better fit calculation between the model and the image, OpenGL provides settings for its magnification (stretch) and minification (shrink) filters.
Each of these filters can behave as a Nearest or a Linear Filter.
As a Nearest filter, the texture coordinates will have a one-to-one correspondence with the texels (the texture-based equivalent of pixels) in the texture map. Whichever texel the coordinate falls in, that color is used for the fragment texture color.
A linear filter however, does not work by taking the nearest pixel, but by applying a weighted average of the texels surrounding the texture coordinate.
Texture Wrapping
UV coordinates fall between the range of [0,1]. However, if UV coordinates fall outside this range, OpenGL handles the coordinates according to the texture wrapping mode. These modes can either be set to Repeat, Clamp, Clamp to edge or Clamp to border.
Loading the image
Finally, we are now able to load the image into the currently bound Texture Object. The loading of the image is done through the OpenGL function glTexImage(). This function requires the raw-image data, height and width of the image. These set of data were produced by the lodepng utility. See method convertImageToRawImage() in Character.mm.
In method setupOpenGL(), go to line 17c and type what is shown in listing 10.
Listing 10. Loading image to the texture buffer
//17c. load the image data into the current bound texture buffer
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, imageWidth, imageHeight, 0,
GL_RGBA, GL_UNSIGNED_BYTE, &image[0]);
Obtaining the location of the Uniform 2DSampler
I have not talked about shaders in depth. I will introduce you to shaders in the next couple of posts. Shaders is a complicated subject and talking about it here may add confusion. For now, absorb how textures are setup and applied to characters on the OpenGL client-side.
Nonetheless, I'm force to talk a bit about shaders and textures. In simple terms, a 2DSampler is a Uniform variable which contains a link to the image loaded using the function glTexImage (see listing 10).
In order to properly address the 2DSampler during rendering, we need to know its location. Our 2DSampler is called TextureMap. (Open the Shader.fsh file if curious to see this Uniform)
We are going to get the location of the 2DSampler by typing what is shown in listing 11. Go to line 18 in the method setupOpenGL and type the following:
Listing 11. Getting the location of the 2DSampler
//18. Get the location of the Uniform Sampler2D
UVMapUniformLocation=glGetUniformLocation(programObject, "TextureMap");
We will use this location in our rendering loop.
Applying the texture to the character
Finally, we are able to apply the texture to the character. This will be done during our rendering loop.
We first activate our texture-unit 0. Once activated, we can bind our texture object. Finally, we link the image in the texture object to the Sampler2D in the shader.
If you recall, we obtained the location of the Sampler2D > > > > TextureMap> > > > in listing 11.
Open up file Character.mm and go to method Draw(). Type lines 3-5 as shown in listing 12.
Listing 12. Applying the texture to the character
//3. Activate the texture unit
glActiveTexture(GL_TEXTURE0);
//4 Bind the texture object
glBindTexture(GL_TEXTURE_2D, textureID[0]);
//5. Specify the value of the UV Map uniform
glUniform1i(UVMapUniformLocation, 0);
Final Result
Run the project. You should now see the robot with its new image apply to it.
Figure 4. A 3D model with texture on a iOS device
Source Code
The source code can be found here.
Questions?
So, do you have any questions? Is there something you need me to clarify? Did this project help you? Please let me know. Add a comment below and subscribe to receive our latest game development projects.
Note:
If you are using a newer Xcode version, you may get the following error:
"Couldn't decode the image. decoder error 29: first chunk is not the header chunk"
If you are getting this error message, the settings in Xcode is preventing the loading of png images.
To fix this, click on the project name, and select "Build Settings". Search for "Compress PNG Files". Set this option (debugger/Release) to NO.
Right below this option, you should see "Remove Text Metadata From PNG FIles". Set this option to NO.
When you build and run the project, the error mentioned above should go away and the png images should show up.
If you need more help, please visit my support page or contact me.
Update:
In newer Xcode versions, you may get this error while running the project demos:
"No such file or directory: ...xxxx-Prefix.pch"
This error means that the project does not have a PCH file. The fix is very simple:
In Xcode, go to new->file->PCH File.
Name the PCH file: 'NameOfProject-Prefix' where "NameOfProject" is the name of the demo you are trying to open. In the OpenGL demo projects, I usually named the projects as "openglesinc."
So the name of the PCH file should be "openglesinc-Prefix"
Make sure the file is saved within the 'NameOfProject' folder. i.e, within 'openglesinc' folder.
Click create and run the project.