公开(公告)号：US20140160355A1

公开(公告)日：2014-06-12

申请号：US13708125

申请日：2012-12-07

Applicant: XEROX CORPORATION

Inventor： Lalit Keshav MESTHA ,  Survi KYAL ,  Sohail A. DIANAT

IPC: H04N5/14

CPC classification number: G06T3/4007 ,  G06T3/4053 ,  H04N7/01 ,  H04N7/0135

Abstract: What is disclosed is a novel video processing system and method wherein a plurality of image frames of a video captured using a video camera with a spatial resolution of (M×N) in the (x, y) direction, respectively, and a temporal resolution (T) in frames per unit of time. A first and second magnification factor f1, f2 are selected for spatial enhancement in the (x, y) direction. A third magnification factor f3 is selected for a desired temporal enhancement in (T). The video data is processed using a dictionary comprising high and low resolution patch cubes which are used to induce spatial and temporal components in the video where no data exists. A high resolution course video X0 is generated which has an enhanced spatial resolution of (f1* M)×(f2*N) and an enhanced temporal resolution of (f3*T) frames. The course high resolution video is then smoothed, when found required, to generate a smoothed high resolution video.

Abstract translation: 所公开的是一种新颖的视频处理系统和方法，其中使用在（x，y）方向上具有（M×N）的空间分辨率的摄像机捕获的视频的多个图像帧分别为（x，y） （T）以每单位时间的帧为单位。 选择第一和第二放大因子f1，f2用于（x，y）方向上的空间增强。 在（T）中为期望的时间增强选择第三放大因子f3。 使用包括高分辨率和低分辨率片段立方体的字典来处理视频数据，这些立方体用于在没有数据存在的视频中诱导空间和时间分量。 产生具有（f1 * M）×（f2 * N）的增强的空间分辨率和（f3 * T）帧的增强的时间分辨率的高分辨率过程视频X0。 当高分辨率视频被发现时需要平滑化，以产生平滑的高分辨率视频。