Free-viewpoint synthesis over panoramic images

PubDate: December 2018

Teams: National Taiwan University，Phillips Exeter Academy

Writers: Tze-How Liew, Yueh-Chun Lai, Hong-Shiang Lin, Sun-Yu Gordon Chi, Ming Ouhyoung

PDF: Free-viewpoint synthesis over panoramic images

Abstract

In the world of virtual reality, the task of constructing a playback data set to navigate through a scene has traditionally required a particularly inefficient procedure. The conventional method of taking pictures and videos with a pinhole camera model is costly due to the slow run time and memory space required. We propose a method which takes advantage of a less costly setup and improves the visual quality of the final images. This method allows users to choose the desired viewpoint, as well as whether the output should be computed as a panoramic or perspective image. This entire procedure consists of four steps: structure from motion (SfM), image rectification and depth estimation, 3D reconstruction, and view synthesis.