Mobile Deployment (Android & iOS)

OpenCV5Sharp features native support for executing computer vision tasks directly on mobile devices (Android and iOS). Because everything executes locally on the device, it provides secure, low-latency execution that works completely offline.

---

📋 Architectural Overview

To run high-performance image processing algorithms on mobile devices, OpenCV5Sharp compiles against native CPU vector extensions:

Platform	Architecture	Vector Backend	Native Library Target
Android	arm64-v8a (64-bit)	ARM NEON	libopencv5sharp_native.so, libopencv_world.so
iOS	arm64 (64-bit)	ARM NEON / Apple Accelerate	libopencv5sharp_native.dylib, libopencv_world.dylib (Framework)

WARNING

Only 64-bit mobile devices and simulators are supported. Attempting to build or run on 32-bit simulators or legacy devices will result in a DllNotFoundException or runtime execution failures.

---

📦 NuGet Installation

To target Android and/or iOS in your .NET MAUI or mobile applications, install the dedicated mobile package:

dotnet add package OpenCV5Sharp.Mobile

---

🤖 Android Integration

To configure your Android application for OpenCV5Sharp:

1. Workload Installation

Ensure that the Android development workload is installed:

dotnet workload install android

2. Permissions Configuration

If you plan to capture live frames using a device camera, you must request permissions in your AndroidManifest.xml (located under Platforms/Android/):

<uses-permission android:name="android.permission.CAMERA" />
<!-- If you plan to save images to the device gallery -->
<uses-permission android:name="android.permission.WRITE_EXTERNAL_STORAGE" />

---

🍎 iOS Integration

To configure your iOS application for OpenCV5Sharp:

1. Workload Installation

Ensure that the iOS development workload is installed:

dotnet workload install ios

2. Code-Signing Requirements

Physical iOS devices require all binaries and native .dylib libraries to be code-signed. The OpenCV5Sharp.Mobile NuGet package includes MSBuild targets that automatically extract, sign, and embed the native interop dynamic libraries into your .app bundle.

• Ensure you have a valid Apple Developer Profile configured in Visual Studio or Rider.
• For physical device debugging, configure your Entitlements.plist and provisioning profiles as you would for a standard iOS app.

---

💾 Resource Management & The Mobile Caching Workaround

The Problem

On Android and iOS, raw files (such as ONNX deep learning models, templates, or camera calibration matrices) embedded in the application package are zipped inside the .apk or .ipa bundle. They do not exist as separate, physical files on the disk.

Because the native C++ OpenCV engine requires direct filesystem path strings (char*) to load assets (such as Cv2.Imread() or Dnn.ReadNet()), it cannot read files directly from inside the compiled bundle.

The Solution: Extracting to Cache

On application startup, you must extract your packaged assets out of the application package and write them to the local device cache directory (FileSystem.CacheDirectory or FileSystem.AppDataDirectory). Once extracted, you can pass the physical path of the cached file to the SDK.

Here is the exact helper method used in our .NET MAUI sample:

using System.IO;
using System.Threading.Tasks;
using Microsoft.Maui.Storage;

public async Task<string> PrepareTempFileFromAssetAsync(string assetName)
{
    // Define the destination target in the local CacheDirectory
    string targetPath = Path.Combine(FileSystem.CacheDirectory, assetName);
    
    // Check if we need to extract it
    if (File.Exists(targetPath))
    {
        File.Delete(targetPath);
    }

    // Open the read stream from the packaged app bundle
    using var stream = await FileSystem.OpenAppPackageFileAsync(assetName);
    
    // Write it to a physical file on disk
    using var outStream = File.OpenWrite(targetPath);
    await stream.CopyToAsync(outStream);
    
    // Return the physical file path that native interop can read
    return targetPath;
}

---

⚡ Performance Optimization Tips

• Free Mat Objects Immediately: Mobile devices have limited memory. Always wrap your Mat manipulations in using blocks to dispose of native memory resources immediately after use.
• Run Heavy Tasks Async: Run video capture loops and DNN inference passes on a background thread using Task.Run to prevent freezing the mobile user interface.
• Isolate Target Workloads: The OpenCV5Sharp NuGet package automatically filters out unnecessary target binaries during compiler staging, keeping your final .apk and .ipa download sizes as small as possible.