I'm interested in putting a view on top of the camera view with React-Native, something similar to this:

Is there a plugin or an approachable way to do this?, for now, I'm only interested in putting an image and recognize a certain person in a room? (4 people are in the same room, and if I point to one, display text A, if I point to other, display text B, and so on)

I'm interested in putting a view on top of the camera view with React-Native, something similar to this:

Is there a plugin or an approachable way to do this?, for now, I'm only interested in putting an image and recognize a certain person in a room? (4 people are in the same room, and if I point to one, display text A, if I point to other, display text B, and so on)

• javascript
• reactjs
• react-native
• react-native-camera

• 1 Last time I checked the plugin react-native-camera was capable of displaying a live image which you could (theoretically) be overlayed with whatever you want. Though I couldn't get the overlay to work on Android. See github./lwansbrough/react-native-camera – FMCorz Commented Jun 9, 2016 at 15:14
• @mitza Sisic did you succeed to do it? – Manspof Commented Oct 15, 2020 at 22:17

1 Answer 1

Certainly, the best camera library is react-native-camera and you can easily use it and style it to wrap all the viewport, see the following codes:

import React, { useRef } from 'react';
import {
  SafeAreaView,
  View,
  Text,
  TouchableOpacity,
  StyleSheet,
} from 'react-native';
import { RNCamera } from 'react-native-camera';

const CameraView = () => {
  const cameraRef = useRef(null);
  const takePicture = async () => {
    if (cameraRef) {
      const options = { quality: 0.5, base64: true };
      const data = await cameraRef.takePictureAsync(options);
      console.log(data.uri);
    }
  };

  return (
    <SafeAreaView style={styles.safeWrapper}>
      <View style={styles.container}>
        <RNCamera
          ref={cameraRef}
          style={styles.camera}
          type={RNCamera.Constants.Type.back}
          flashMode={RNCamera.Constants.FlashMode.on}
        />
        <View style={styles.snapWrapper}>
          <TouchableOpacity onPress={takePicture} style={styles.capture}>
            <Text style={styles.snapText}>SNAP</Text>
          </TouchableOpacity>
        </View>
      </View>
    </SafeAreaView>
  );
};

const styles = StyleSheet.create({
  safeWrapper: {
    flex: 1,
  },
  container: {
    flex: 1,
    flexDirection: 'column',
    backgroundColor: 'black',
    position: 'relative',
  },
  camera: {
    flex: 1,
    justifyContent: 'flex-end',
    alignItems: 'center',
  },
  capture: {
    flex: 0,
    backgroundColor: '#fff',
    borderRadius: 5,
    padding: 15,
    paddingHorizontal: 20,
    alignSelf: 'center',
    margin: 20,
  },
  snapWrapper: {
    flex: 0,
    flexDirection: 'row',
    justifyContent: 'center',
    backgroundColor: 'rgba(0, 0, 0, 0)',
    position: 'absolute',
    top: 50,
    left: 16,
    right: 16,
  },
  snapText: {
    fontSize: 14,
    color: 'red',
  },
});

export default CameraView;

View of the above code:

Face Detection

For face detection there is onFacesDetected that you can use it like below:

<RNCamera
  ref={cameraRef}
  style={styles.camera}
  type={RNCamera.Constants.Type.back}
  flashMode={RNCamera.Constants.FlashMode.on}
  onFacesDetected={({ faces }) => { // <===== this function
    console.log(faces); 
  }}
/>

And faces type is:

interface Size<T = number> {
  width: T;
  height: T;
}

interface Point<T = number> {
  x: T;
  y: T;
}

interface Face {
  faceID?: number;
  bounds: {
    size: Size;
    origin: Point;
  };
  smilingProbability?: number;
  leftEarPosition?: Point;
  rightEarPosition?: Point;
  leftEyePosition?: Point;
  leftEyeOpenProbability?: number;
  rightEyePosition?: Point;
  rightEyeOpenProbability?: number;
  leftCheekPosition?: Point;
  rightCheekPosition?: Point;
  leftMouthPosition?: Point;
  mouthPosition?: Point;
  rightMouthPosition?: Point;
  bottomMouthPosition?: Point;
  noseBasePosition?: Point;
  yawAngle?: number;
  rollAngle?: number;
}

After facing detection and get the faces response you can use state and add a new view design and showing on the view that I exampled on the top of the post.