Method for generating 3D images in all directions using 2D images of objects based on machine learning based on AI

摘要

The method for generating a stereoscopic (3D) image in all directions with a unidirectional (2D) image of an AI-based machine learning-based object according to an embodiment of the present invention is that the second server receives the object image, the actual size of the object and the actual main object from the first server. A step of receiving location information, performing AI image learning on the information, and generating and storing a weight that creates an image in a specific direction and a specific viewpoint distance as a result of the AI image learning, by the second server from the first server Receiving the classification of the object to be created, the specific direction, and the location information at a specific time, the second server searching for the weight corresponding to the specific direction and the specific time of the object, and generating and storing the image in the desired direction; 2 It may include the step of creating a video (VOD file, etc.) by connecting the generated images in the server.