Beyond fetching and parsing the binary representation of the face detection model and setting it on the state, we’re creating a new camvas, which takes a reference to a <canvas> context and a ...