README-en.md

What is fusion?

fusion is an OpenGL ES effects rendering library on Android written by kotlin, which is similar to GPUImage on IOS.

• fusion highly abstracts the input / output and rendering process, hides the complex and trivial OpenGL API. Developers without OpenGL knowledge can use fusion easily
• organize and manage the renderers by chain/graph through RenderChain / RenderGraph, and unify input and output through RrenderPipline.
• support image/video/camera as input, and video decode/encode, camera logic.
• support image/video offscreen rendering to save the effects.
• support auto reuse texture/frame buffer/program to reduce the usage of graphic memory.
• support GL thread and EGL to easily create GL enviroment.
• support display the rendered result by the built-in view, and also support GLSurfaceView .
• supports common rendering effects, and developers can inherit SimpleRenderer/RenderChain/RenderGraph to achieve complex effects, or implement Renderer interface.

continuously updating...

Integration：

add the following code to the root gradle of your project:

allprojects {
    repositories {
    	...
    	maven { url 'https://www.jitpack.io' }
    }
}

then add the following code to the gradle of your module:

dependencies {
	implementation 'com.github.kenneycode:fusion:1.2.0'
}

Basic usage of image rendering

// create RenderChain and add some renderers
val renderer = RenderChain()
	.addRenderer(ScaleRenderer().apply { setFlip(false, true); setScale(0.8f) })
	.addRenderer(CropRenderer().apply { setCropRect(0.1f, 0.9f, 0.8f, 0.2f) })
	.addRenderer(LUTRenderer().apply { setLUTImage(Util.decodeBitmapFromAssets("test_lut.png")!!); setLUTStrength(0.8f) })
	.addRenderer(GaussianBlurRenderer().apply { setBlurRadius(10) })

// create RenderPipeline，connecting input, renderer and ouput
renderPipeline = RenderPipeline
	.input(FusionImage(Util.decodeBitmapFromAssets("test.png")!!))
	.renderWith(renderer)
	.useContext(fusionView)
	.output(fusionView)

// start processing
renderPipeline.start()

Basic usage of video rendering

// create RenderChain and add some renderers
val renderer = RenderChain()
	.addRenderer(OES2RGBARenderer())
	.addRenderer(LUTRenderer().apply { setLUTImage(Util.decodeBitmapFromAssets("test_lut.png")!!); setLUTStrength(0.8f) })
	.addRenderer(GaussianBlurRenderer().apply { setBlurRadius(10) })

// create RenderPipeline，connecting input, renderer and ouput
renderPipeline = RenderPipeline
    .input(FusionVideo("/sdcard/test.mp4"))
    .renderWith(renderer)
    .useContext(fusionView)
    .output(fusionView)

// start processing
renderPipeline.start()

Basic usage of camera rendering

// create RenderChain and add some renderers
val renderer = RenderChain()
	.addRenderer(OESConvertRenderer())
	.addRenderer(LUTRenderer().apply { setLUTImage(BitmapUtil.decodeBitmapFromAssets("test_lut.png")!!); setLUTStrength(0.8f) })

// configs for constructing fusion camera
val fusionCameraConfig = FusionCamera.Config().apply {
    windowRotation = activity!!.windowManager.defaultDisplay.rotation
    desiredPreviewSize = Size(1080, 1920)
}

// create RenderPipeline，connecting input, renderer and ouput
renderPipeline = RenderPipeline
    .input(FusionCamera(fusionCameraConfig))
    .renderWith(renderer)
    .useContext(fusionView)
    .output(fusionView)

// start processing
renderPipeline.start()

for more usages please see the demo.

thank you!