METHOD AND APPARATUS OF CONVERTING 3D VIDEO IMAGE FROM VIDEO IMAGE USING DEEP LEARNING

摘要

In a method of converting a video image into a 3D video image, a training video image is captured using at least two cameras. A 3D modeled training 3D video image is generated using the training video image, and an artificial neural network model is trained using the training video image and the training 3D video image. A video image to be converted is input to the learned artificial neural network model. An image is transformed to generate a 3D video image from the video image using the learned artificial neural network model. Output to output the 3D video image.