Integration

This chapter will introduce the integration methods for GPUPixel across different systems. The latest pre-compiled libraries for each platform can be downloaded here.

iOS & MacOS

The linking method for both iOS and MacOS libraries is the same, so they will be covered together.

Adding Dependencies

Copy the gpupixel and vnn libraries to your project directory.

iOS

├── gpupixel.framework
├── vnn_core_ios.framework
├── vnn_face_ios.framework
└── vnn_kit_ios.framework

MacOS

├── gpupixel.framework
├── vnn_core_osx.framework
├── vnn_face_osx.framework
└── vnn_kit_osx.framework

Select Project -> Targets -> Build Phases -> Link Binary With Libraries.

Click the + button in the lower left corner to add the dependencies mentioned above. The CoreMedia.framework and AVFoundation.framework are system libraries that provide capabilities such as camera capture. After adding them, the situation should resemble the following image.

Framework Search Paths

Click Targets -> Build Settings -> Search Paths -> Framework Search Paths, input gpupixel和vnn framework paths.

Embedding Frameworks

Select Project -> Targets -> General -> Frameworks, Libraries, and Embedded Content.

Set gpupixel and vnn libraries to Embed & Sign, choose Do Not Embed for system libraries, as shown below.

Objective-C Invocation

Since GPUPixel is written in C++, to call C++ from Objective-C (OC), the file extension for OC source files must be changed from .m to .mm. This allows for mixed compilation between OC and C++.

All public headers for GPUPixel are included in gpupixel.h. By importing

#import <gpupixel/gpupixel.h>

you can access all the functions. For detailed library function calls, refer here.

Swift Invocation

Starting from Swift 5.9, direct interaction with C++ is possible, also through bridging header files, eliminating the need for an intermediate step with Objective-C. You can refer to this article or the official Swift documentation on Mixing Swift and C++.

Android

Download the latest gpupixel-release.aar and place it under your Android project, such as in the libs directory.

Add the dependency in the gradle file:

dependencies {
    implementation files('libs/gpupixel-release.aar')
}

Then sync the project. Refer to the Android Demo for more details.

Reference article: Importing AAR Package in Android

Windows

Linux