The overview of 2D to 3D conversion system

摘要

3D signal processing has become a hot trend in the related visual processing field. But at this stage, the lack of 3D content is becoming an important factor that limits its development. It will be great help to solve this problem if we extract the depth information from the monocular image and automatically convert it to 3D video. An automatic 2D-to-3D conversion will greatly reduce the production cost, while providing a richer and higher quality 3D content. On the basis of some research findings in recent years, this paper will introduce the main two kinds of methods to do the conversion: off-line conversion and real-time conversion. Taking off-line conversion for example, we systematically presents all aspects of the conversion process, including the various methods that each aspect can use, then we describe the principles and basic methods for real-time conversion, and finally discuss the challenge and development that 2D to 3D image conversion technology will be in the future.