Abstract

Digital 3D environments are crucial in Virtual Reality, because they can be used in various applications such as virtual Job Training. However, creating 3D scenes manually is time consuming. If computers are enabled to generate 3D environments, it can save huge effort on different occasions. Many approaches for 3D scene reconstruction from a RGB image have been proposed. However, a significant number of them use perspective images as input, which contain less field of view and contextual information compared to 360 panorama images. Also, with continuous academic advancement of Convolutional Neural Network(CNN), it can be heavily utilized to estimate different aspects during the scene reconstruction process, with a relatively high accuracy. The goal of the research is to estimate complete indoor rooms that feature the room layout, and are populated with CAD models objects that are present in the image, positioned and rotated accordingly to the image. The final visualization of the environment will be displayed in the popular game engine, Unity. In the presentation, related works, as well as my proposed approach, will be introduced. After that, the difference between previous and the proposed method will be explained, and future works will be discussed