Kiana Calagari, Piotr Didyk, Mohamed Elgharib, Alexandre Kaspar, Mohamed Hefeeda and Wojciech Matusik
ACM Multimedia, 2015
A wide spread adoption of 3D stereoscopic television is hindered by the lack of high-quality 3D content. One promising solution to address this need is to use automated 2D-to-3D conversion. Current conversion methods, while general, produce low-quality results that exhibit artifacts that are not acceptable to many viewers. We address this problem by showing how to construct a high-quality, domain-specific conversion method for sports videos. Our method is data-driven — we generate stereoscopic frames for a given video stream by transferring depth information from a database of similar 3D stereoscopic videos. Once we retrieve similar 3D video frames, our technique transfers depth gradients to the target frame while respecting object boundaries, solves for the output depth map, and then generates stereoscopic video.