AVCaptureVideoPreviewLayer: pointForCaptureDevicePointOfInterest() method
The pointForCaptureDevicePointOfInterest() method converts a 0-to-1 fractional coordinate in the picture into a point in this layer. It's the reverse of captureDevicePointOfInterestForPoint().
Syntax
pointForCaptureDevicePointOfInterest(point)Parameters
point
A point in the picture, as fractions from 0 to 1; see Coordinates in the picture. For example, the camera's current focusPointOfInterest. It can be a CGPoint or a plain object with the two numbers x and y.
Return value
A CGPoint: a point in this layer, in pt, with the origin at the layer's top-left corner.
Verified by testing: while the session is capturing and the picture fills a 180 × 240 pt layer, (0.5, 0.5) converts back to (90, 120), off by less than 1 pt.
The result depends on the session's current picture; compute it after the session has started capturing. Values computed before capturing starts are not verified on a device.
Exceptions
TypeError
pointisn't an object:pointForCaptureDevicePointOfInterest needs an object with all of: x, y.- A field is missing or isn't a finite number:
pointForCaptureDevicePointOfInterest is missing …, or it is not a finite number.
Examples
const camera = AVCaptureDevice.defaultDeviceWithMediaType('video');
if (camera !== null && AVCaptureDevice.authorizationStatusForMediaType('video') === 'authorized') {
const session = new AVCaptureSession();
session.addInput(new AVCaptureDeviceInput(camera));
const preview = new AVCaptureVideoPreviewLayer(session);
preview.videoGravity = 'resizeAspectFill';
preview.frame = new CGRect(0, 0, 180, 240);
session.startRunning().then(() => {
const point = preview.pointForCaptureDevicePointOfInterest({ x: 0.5, y: 0.5 });
console.log(Math.round(point.x), Math.round(point.y));
});
}Effect: When access is granted, after the session starts capturing, the console prints 90 120: the center of the picture is at the center of the layer.
Corresponding iOS API
-[AVCaptureVideoPreviewLayer pointForCaptureDevicePointOfInterest:].
See also
captureDevicePointOfInterestForPoint(): convert from the layer to the picturefocusPointOfInterest