公开(公告)号：WO2015138622A1

公开(公告)日：2015-09-17

申请号：PCT/US2015/019992

申请日：2015-03-11

Applicant: AMAZON TECHNOLOGIES, INC.

Inventor： HEINZ, II, Gerald Joseph ,  PESCE, Michael Schleif ,  DAVIS, Collin Charles ,  FRAZZINI, Michael Anthony ,  ALKARMI, Ashraf ,  GEORGE, Michael Martin ,  LIMP, David A. ,  CARR, JR., William Dugald

IPC: H04N7/16

CPC classification number: H04N21/47205 ,  G06T19/20 ,  H04N21/431

Abstract: Real-time video targeting and exploration system that allows users to pause, step into, and explore modeled worlds of scenes in video. The system may leverage network-based computation resources to render and stream new video content from the models to clients with low latency. A user may pause a video, step into a scene, and interactively change viewing positions and angles to move through or explore the scene, as well as interactively explore objects associated with the scene. The user may step into and explore the scene within the scope of the model to discover parts of the scene that are not visible in the original video, as well as objects within the scene that may not have been readably observable in the original video. In addition, at least some content of a video may be replaced with content targeted at particular viewers according to viewers' profiles or preferences.

Abstract translation: 实时视频定位和探索系统，允许用户暂停，进入和探索视频中场景的建模世界。 该系统可以利用基于网络的计算资源，以低延迟将来自模型的新视频内容呈现给客户端。 用户可以暂停视频，进入场景，并交互地改变观看位置和角度以移动或探索场景，以及交互地探索与场景相关联的对象。 用户可以进入和探索模型范围内的场景，以发现在原始视频中不可见的场景的部分以及场景中可能在原始视频中可以被可读地观察到的对象。 此外，根据观看者的个人资料或偏好，视频的至少一些内容可以被替换为针对特定观看者的内容。