Exploration and Analysis of Various Techniques used in 2D to 3D Image and Video Reconstruction

This review article provides an abstract comparison of various popular methods of 2D to 3D image and/or video conversion. Even though 3D hardware is becoming increasingly widespread, 3D content remains dominated by its 2D cousin. Some of the techniques use deep learning some on the other hand have human-based interaction to convert 2d images into 3d models. Here some techniques are automatic while other are semi-automatic. The major strategies that have lately been popular include gathering numerous photos from varying distances and feeding them to a model to generate a sparse collection of 3D points via triangulation based on feature points seen in the photographs.