-
公开(公告)号:US08649558B2
公开(公告)日:2014-02-11
申请号:US13149609
申请日:2011-05-31
申请人: Michael Gleicher , Feng Liu , Yuzhen Niu
发明人: Michael Gleicher , Feng Liu , Yuzhen Niu
IPC分类号: G06K9/00
CPC分类号: G06T3/0093
摘要: Video data is processed with region-based warping. In accordance with various example embodiments, video data having a plurality of source video frames is processed as follows, for each source video frame. Pixels representing the source video frame and at least one temporally-adjacent video frame are used to generate a motion history map, based upon frame-to-frame movement of at least one region in the source video frame and temporally-adjacent video frames. At least one salient region is identified in the source video frame based upon the motion history map and static image saliency data. Weighting data is determined for warping pixels corresponding to background regions of the source video frame differently than pixels corresponding to the at least one salient region in the source video frame. Using the weighting data, the pixels representing the source video frame are warped to generate pixels representing a warped video frame.
摘要翻译: 视频数据采用基于区域的翘曲进行处理。 根据各种示例实施例,对于每个源视频帧,如下处理具有多个源视频帧的视频数据。 表示源视频帧和至少一个时间相邻视频帧的像素被用于基于源视频帧和时间相邻视频帧中的至少一个区域的帧到帧移动来生成运动历史图。 基于运动历史图和静态图像显着性数据,在源视频帧中识别出至少一个显着区域。 确定与源视频帧的背景区域对应的翘曲像素的加权数据,不同于与源视频帧中的至少一个显着区域对应的像素。 使用加权数据,表示源视频帧的像素被扭曲以产生表示翘曲视频帧的像素。
-
公开(公告)号:US20120307155A1
公开(公告)日:2012-12-06
申请号:US13149609
申请日:2011-05-31
申请人: Michael Gleicher , Feng Liu , Yuzhen Niu
发明人: Michael Gleicher , Feng Liu , Yuzhen Niu
IPC分类号: H04N9/74
CPC分类号: G06T3/0093
摘要: Video data is processed with region-based warping. In accordance with various example embodiments, video data having a plurality of source video frames is processed as follows, for each source video frame. Pixels representing the source video frame and at least one temporally-adjacent video frame are used to generate a motion history map, based upon frame-to-frame movement of at least one region in the source video frame and temporally-adjacent video frames. At least one salient region is identified in the source video frame based upon the motion history map and static image saliency data. Weighting data is determined for warping pixels corresponding to background regions of the source video frame differently than pixels corresponding to the at least one salient region in the source video frame. Using the weighting data, the pixels representing the source video frame are warped to generate pixels representing a warped video frame.
摘要翻译: 视频数据采用基于区域的翘曲进行处理。 根据各种示例实施例,对于每个源视频帧,如下处理具有多个源视频帧的视频数据。 表示源视频帧和至少一个时间相邻视频帧的像素被用于基于源视频帧和时间相邻视频帧中的至少一个区域的帧到帧移动来生成运动历史图。 基于运动历史图和静态图像显着性数据,在源视频帧中识别出至少一个显着区域。 确定与源视频帧的背景区域对应的翘曲像素的加权数据,不同于与源视频帧中的至少一个显着区域对应的像素。 使用加权数据,表示源视频帧的像素被扭曲以产生表示翘曲视频帧的像素。
-
公开(公告)号:US07574069B2
公开(公告)日:2009-08-11
申请号:US11194804
申请日:2005-08-01
申请人: Vidya Setlur , Michael Gleicher , Bruce Gooch , Saeko Takagi , Ramesh Raskar
发明人: Vidya Setlur , Michael Gleicher , Bruce Gooch , Saeko Takagi , Ramesh Raskar
IPC分类号: G06K9/36
CPC分类号: G06K9/3233 , G06T3/0012 , G06T5/005 , G06T7/11 , G06T7/162 , G06T11/60 , H04N1/3875
摘要: A method retargets an image to a different size. An input image is segmented into regions. Selected regions are cut from the input image to construct a background image. The background image is scaled to a predetermined size, and the selected regions are pasted back into the scaled background image to produce an output image.
摘要翻译: 一种方法将图像重定向到不同的大小。 输入图像被分割成区域。 从输入图像切割所选择的区域以构建背景图像。 将背景图像缩放到预定大小,并且将所选择的区域粘贴回缩放的背景图像以产生输出图像。
-
公开(公告)号:US20070025637A1
公开(公告)日:2007-02-01
申请号:US11194804
申请日:2005-08-01
申请人: Vidya Setlur , Michael Gleicher , Bruce Gooch , Saeko Takagi , Ramesh Raskar
发明人: Vidya Setlur , Michael Gleicher , Bruce Gooch , Saeko Takagi , Ramesh Raskar
CPC分类号: G06K9/3233 , G06T3/0012 , G06T5/005 , G06T7/11 , G06T7/162 , G06T11/60 , H04N1/3875
摘要: A method retargets an image to a different size. An input image is segmented into regions. Selected regions are cut from the input image to construct a background image. The background image is scaled to a predetermined size, and the selected regions are pasted back into the scaled background image to produce an output image.
摘要翻译: 一种方法将图像重定向到不同的大小。 输入图像被分割成区域。 从输入图像切割所选择的区域以构建背景图像。 将背景图像缩放到预定大小,并且将所选择的区域粘贴回缩放的背景图像中以产生输出图像。
-
-
-