发明名称 Multi-perspective stereoscopy from light fields
摘要 Methods and systems for generating stereoscopic content with granular control over binocular disparity based on multi-perspective imaging from representations of light fields are provided. A reference image, a three-dimensional (“3D”) representation of a light field corresponding to the reference image, and a goal disparity image that indicates a goal binocular disparity for one or more pixels of the reference image may be received. For each pixel of an output image corresponding to the reference image, a point within the light field that is a closest match for the goal binocular disparity of a corresponding pixel of the goal disparity image may be determined. A stereoscopic image pair including the reference image and the output image may be generated.
申请公布号 US9165401(B1) 申请公布日期 2015.10.20
申请号 US201414156368 申请日期 2014.01.15
申请人 DISNEY ENTERPRISES, INC.;ETH ZÜRICH (EIDGENÖESSISCHE TECHNISCHE HOCHSCHULE ZÜRICH) 发明人 Kim Changil;Hornung Alexander Sorkine;Mueller Ulrich;Pritch Yael;Zimmer Henning;Gross Markus
分类号 G06T15/50 主分类号 G06T15/50
代理机构 Kilpatrick Townsend & Stockton LLP 代理人 Kilpatrick Townsend & Stockton LLP
主权项 1. A computer-implemented method for generating stereoscopic content from a three-dimensional (3D) representation of a light field representing virtual or actual light rays from a scene, wherein the 3D representation is readable in electronic form, the method comprising: receiving a 3D representation of the light field; obtaining a reference image that is included in the 3D representation of the light field; receiving a goal disparity image that indicates a goal binocular disparity for one or more pixels of the reference image; generating a dual reference image by shifting the reference image using the goal disparity image; for each pixel of an output image corresponding to the reference image: determining a point within the light field that is a closest match for the goal binocular disparity of a corresponding pixel of the goal disparity image by optimizing an energy function that compares the dual reference image and the 3D representation of the light field wherein the energy function comprises a sum of a data term and a smoothness term, wherein the data term calculates a normed difference of the dual reference image and the light field; and generating, using the computer, a stereoscopic image pair comprising the reference image and the output image.
地址 Burbank CA US