React Native

Props

# VisionCamera Component Props

The VisionCamera component is the primary camera component for barcode scanning and OCR. It provides a streamlined API without requiring API keys or cloud configuration for basic scanning functionality.

# Detection Config Object

        detectionConfig={{
  text: true,              // Enable text detection (iOS only)
  barcode: true,           // Enable barcode detection
  document: true,          // Enable document detection (iOS only)
  sharpness: false,        // Enable live image sharpness scoring (default false, opt-in since it costs extra CPU/Neural Engine work)
  barcodeConfidence: 0.5,  // Barcode detection confidence (0-1, iOS only)
  documentConfidence: 0.5, // Document confidence (0-1, iOS only)
  documentCaptureDelay: 2.0 // Delay before auto-capture (seconds, iOS only)
}}

      

# VisionCameraRefProps — Imperative Methods Reference

# Pausing and Resuming Detection

pauseDetection() and resumeDetection() let you temporarily stop all detection work (barcode, text, document, and sharpness) without tearing the camera down:

        // Stop detection, e.g. right after a successful capture while showing a loading state
cameraRef.current?.pauseDetection();

// Resume detection later, e.g. once the loading state is dismissed
cameraRef.current?.resumeDetection();

      

This is different from calling stop(). pauseDetection() keeps the live preview visible and the camera session active, so resuming is instant, with no camera restart flicker and no permission or session setup work needed again. Use it for short-lived UI states (a loading spinner, a result screen) where you don't want the camera analyzing frames in the background, but don't want to stop the camera entirely either.

While paused:

• None of the detection event callbacks fire (onRecognitionUpdate, onBarcodeDetected, onBoundingBoxesUpdate, onSharpnessScoreUpdate).
• Any bounding boxes or detection indicators currently on screen are cleared immediately.
• Any in-progress price tag detection is cancelled rather than left to finish in the background.

Switching scanMode automatically resumes detection even if it was previously paused, since changing modes rebuilds the underlying native scanner. If your UI has its own pause toggle, resync it to "on" when the user changes scan mode, the same way the example app does.

# VisionCore Methods

For camera-independent OCR prediction and logging operations, use the VisionCore module instead of the component ref methods:

• Model Management: Use VisionCore.downloadModel(), VisionCore.loadOCRModel(), etc. — See Documentation
• Headless OCR Prediction: Use VisionCore.predictWithModule(), VisionCore.predictShippingLabelCloud(), etc. — See Documentation
• Data Logging: Use VisionCore.logItemLabelDataToPx(), VisionCore.logShippingLabelDataToPx() — See Documentation
• Camera Capabilities: Use VisionCore.getCameraCapabilities() to snapshot the device's lenses, zoom stops, and torch/focus support without mounting a camera - See Documentation

        import { VisionCore } from 'react-native-vision-sdk';

// Model management (on-device OCR)
const module = { type: 'shipping_label', size: 'large' };
await VisionCore.downloadModel(module, apiKey, token, progressCallback);
await VisionCore.loadOCRModel(module, apiKey, token);

// Camera-independent prediction
const result = await VisionCore.predictWithModule(module, imageUri, barcodes);

// Or cloud prediction (no model required)
const cloudResult = await VisionCore.predictShippingLabelCloud(imageUri, [], options);

// Data logging
await VisionCore.logItemLabelDataToPx(imageUri, barcodes, responseData, token, apiKey, true, metadata);