AI Video Processor SDK: Universal Android Video Effects 🚀

The AI Video Processor SDK is a high-performance video augmentation library for Android. It allows you to add professional AI effects to your video streams with just a few lines of code.

---

🌟 Key Features

• Universal Support: Works with WebRTC, CameraX, Camera2, and custom video pipelines.
• Real-Time AI Effects:
  • Background Blur: High-performance bokeh effects.
  • Virtual Backgrounds: Replace backgrounds with images or colors.
  • Smart Watermarking: Precise logo and text overlays.
• Zero-Lag Performance: Native libyuv C++ engine ensures maximum FPS and low battery drain.
• Plug-and-Play: Optimized default settings so you can start immediately.

---

📦 Installation

Add JitPack to your settings.gradle or root build.gradle:

dependencyResolutionManagement {
    repositoriesMode.set(RepositoriesMode.FAIL_ON_PROJECT_REPOS)
    repositories {
        google()
        mavenCentral()
        maven { url = uri("https://jitpack.io") }
    }
}

Add the dependency to your app's build.gradle:

dependencies {
    implementation("com.github.codewithkael:AiVideoProcessor:latest-tag")
}

---

🛠 Integration Guide

1. WebRTC Integration (Simplest)

If you are using WebRTC, use the AiVideoCapturerObserver. It handles all the complex frame conversions and AI processing for you automatically.

// 1. Define your effects
val config = AiVideoProcessorConfig(
    blurBackground = BlurBackgroundConfig(enabled = true, blurRadius = 15f)
)

// 2. Wrap your observer
val aiObserver = AiVideoCapturerObserver(
    context = applicationContext,
    downstream = myOriginalObserver, // The observer you already have
    config = config
)

// 3. Use it in your video capturer
videoCapturer.initialize(surfaceTextureHelper, context, aiObserver)

2. CameraX / Custom Camera Integration

For non-WebRTC features, use the FrameProcessor.

Note

What is helpers? The helpers object provides the library with access to the AI engine (ML Kit) and drawing tools. You can create it with a single line using the built-in factory.

// 1. Create the default engine helpers (Only needed once)
val helpers = FrameProcessingHelpers.default(applicationContext)

// 2. Initialize the processor
val processor = FrameProcessor.fromConfig(config, helpers)

// 3. In your analysis loop (e.g., CameraX ImageAnalysis)
val processedBitmap = processor.process(inputBitmap)

3. Updating Filters in Real-Time

You can change settings (like turning on a watermark or changing the blur radius) while the camera is running.

val newConfig = currentConfig.copy(
    watermark = WatermarkConfig(enabled = true, text = "Property of Company")
)

// The change is applied instantly to the video stream
aiObserver.updateConfig(newConfig) 

---

⚙️ Configuration Reference

AiVideoProcessorConfig

This is the main object you use to turn features on or off.

Property	Description
blurBackground	Control the bokeh/blur intensity.
replaceBackground	Provide an image to use as a virtual background.
watermark	Add a logo or text to the corner of the video.

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.