ABSTRACT
This chapter presents a literature survey that concentrates on converting conventional monocular video sequences, or those acquired by one camera, to produce two or more views of the same scene, to be presented for viewing using the end three-dimensional (3D) technology. It discusses the relevant and two-dimensional (2D) to 3D methods by using motion estimation as the underlying framework. The chapter explains the 2D to 3D methods using scene analysis, or other features extracted from the scene. It summarizes the methods in each category, the disadvantages and drawbacks that can be encountered, leading to what can be formulated for future research in this area. Features such as shape, edges, color, or anything involving the direct use of features other than motion are of interest. Interestingly, some approaches to 2D to 3D conversion use shape and texture features to generate the stereoscopic content.
