How do I best implement an AR tool that would allow me to display pictures onto a sized frame on the wall? I'm currently using Canny edge detection and finding a shape with 4 corners, then warping the perspective of an input image for each frame of a video.
Is there a better way to do this? I've experimented with Aruco markers but would prefer a way that does not require camera calibration, and there is no need for pose estimation as well.