A Swift camera framework for iOS that simplifies AVFoundation usage and integrates CoreML models for real-time object recognition.
Lumina is a Swift camera framework for iOS that simplifies integrating camera functionality and CoreML models into applications. It provides a ready-to-use camera module that handles photo/video capture, live streaming, QR/barcode scanning, and real-time object recognition without requiring developers to write low-level AVFoundation code.
iOS developers building applications that require camera functionality, especially those integrating machine learning models for real-time image recognition or needing advanced features like depth data capture and metadata detection.
Lumina saves significant development time by abstracting complex AVFoundation code, offering a unified solution for camera operations and CoreML integration with a customizable UI and extensive feature set out of the box.
A camera designed in Swift for easily integrating CoreML models - as well as image streaming, QR/Barcode detection, and many other features
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Streams real-time object recognition from any CoreML-compatible model directly in the camera feed, as shown with MobileNet and SqueezeNet in the sample app, eliminating separate model integration code.
Supports capturing still images, videos, live photos, and depth data in a single framework, reducing the need for multiple camera implementations and boilerplate AVFoundation code.
Allows showing/hiding built-in buttons or embedding the view controller in custom interfaces, demonstrated with code snippets for external controls and UI adaptation.
Scans QR/barcodes and detects faces automatically, outputting metadata without additional setup, streamlining apps that require these features.
Requires iOS 13.0 or higher, limiting support for older devices and apps with broader compatibility needs, as stated in the requirements.
Certain features like depth data capture and streaming require specific resolution settings, and video recording mode disables frame streaming, adding configuration complexity and potential pitfalls.
Relies on swift-log under Apache 2.0 license alongside Lumina's MIT license, which may complicate licensing management for projects with strict policy requirements.