An effective image distortion correction method using depth sensor

Abstract

We utilize Kinect v2, which is a sensor deeply, to provide a way to recognize walls and interpolate mapping images accordingly. It recognizes the angle and shape of the wall and basically interpolates the recognized form. First, we will use depth sensors to receive wall depth data. The data received in this way is converted to primary usable data via preprocessing. By using subsequent deformed data, we calculate by deducing the extent and degree of coking of the wall. Using the calculated result, Unity 3D will correct the image and interpolate. We show that it is possible to propose a method which can more easily map through this process through research and experiment.