Closed joratapok closed 1 year ago
I could indeed pass the frame as a second argument to the onBarcodeScanned callback so you can access it easily.
Meanwhile you can also write your own frame processor leveraging scanCodes
to access the frame directly, eg. https://github.com/mgcrea/vision-camera-barcode-scanner/issues/5#issuecomment-1761004490.
Regarding the barcode bonding boxes they are quite messy and need pretty complex translations to actually match the camera frame, it is something that is done to compute highlights but I guess I should do it by default in the returned object so that you get useable coordinates outside the box.
I'm trying to filter barcodes that only in square layout, but I don't have frame size, and position in barcodes do not equal screen size or screen resolution