react-native-nitro-vision-kit
A React Native library for on-device subject cutouts, image labeling and OCR, using Apple's Vision framework on iOS and Google's ML Kit on Android through Nitro Modules.
react-native-nitro-vision-kit exposes on-device computer vision to React Native applications: cutting a subject out of a photo with an optional mask, ranking labels for what an image contains, and reading text in several scripts, all processed offline on the device rather than through a cloud API. A single decode can run all three operations together for efficiency.
It is built with Nitro Modules for the native bridge, requires React Native's New Architecture, and needs a minimum of React Native 0.75, iOS 13 with cutouts requiring iOS 17 and OCR requiring iOS 18, or Android API 24. It installs through npm alongside its Nitro Modules dependency, with a CocoaPods install step on iOS.