首页> 外国专利> APPARATUS AND A METHOD FOR MIXING THREE-DIMENSIONAL GRAPHIC IMAGE IN SPORTS VIDEO TO MAKE A VIEWER OF THE SPORTS VIDEO FEEL THAT THE GRAPHIC IMAGE REALLY EXISTS IN STADIUM

APPARATUS AND A METHOD FOR MIXING THREE-DIMENSIONAL GRAPHIC IMAGE IN SPORTS VIDEO TO MAKE A VIEWER OF THE SPORTS VIDEO FEEL THAT THE GRAPHIC IMAGE REALLY EXISTS IN STADIUM

机译:用于在体育视频中混合三维图形图像的设备和方法,以使观看者感觉到体育图形中确实存在图形图像

摘要

PURPOSE: An apparatus and a method for mixing 3D(three-Dimensional) graphic image in sports video are provided to make viewer of the sports video feel that the 3D graphic image really exists in a stadium by mixing a graphic image with the sports video at its desired location. CONSTITUTION: A corner point, used for searching cross information of line components, is calculated on a basis of model information of a standardized stadium. A graphic object is mixed with sports video by using information about the corner point. An affined transform and a perspective transform are calculated by utilizing the information about the corner point. A graphic image which is supposed to be mixed with the sports video is transformed by using information of the affined transform and the perspective transform. A plurality of corner points are detected continuously from a video frame. On a basis of information about a previous frame, locations of the corner points in a next frame are pre-estimated.
机译:目的:提供一种用于在体育视频中混合3D(三维)图形图像的设备和方法,以通过在体育视频中将图形图像与体育视频混合,使体育视频的观看者感觉到3D图形图像确实存在于体育场内其所需的位置。构成:用于搜索线成分的交叉信息的拐角是根据标准化体育场的模型信息计算得出的。通过使用有关拐角点的信息将图形对象与体育视频混合。通过利用关于拐角点的信息来计算仿射变换和透视变换。通过使用仿射变换和透视变换的信息来变换应该与体育视频混合的图形图像。从视频帧连续检测出多个拐角点。基于关于前一帧的信息,下一帧中拐角点的位置被预先估计。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号