Abstract: A method and related system of encoding or re-encoding video is disclosed. In the method, video data comprising a sequence of frames is received. For a respective frame in the sequence of frames, a multi-level frame is generated comprising the respective frame and a plurality of copies of the respective frame. Each copy has an associated video resolution level that is a member of a predefined range of video resolution levels, ranging from a highest video resolution level to a lowest video resolution level. The multi-level frame is encoded.
Abstract: A method of displaying video includes receiving a video bitstream corresponding to a video. The video comprises a sequence of frames having an associated video resolution level. At least a portion of the video bitstream is decoded. A background region of the video is displayed at a first video resolution level and a window region of the video is displayed at a second video resolution level. The second video resolution level is higher than the first video resolution level.
Type:
Grant
Filed:
June 24, 2008
Date of Patent:
August 7, 2012
Assignee:
Precoad Inc.
Inventors:
Charles K. Chui, Haishan Wang, Dongfang Shi
Abstract: A method of filtering a digital image is described. A filter kernel is applied to a respective pixel in a set of pixels to smooth noise and preserve spatial frequencies associated with image edges in the digital image in accordance with a first filtering parameter. The filter kernel is a function of the respective pixel and has a closed form for the respective pixel. The filter kernel includes contributions from a first set of neighboring pixels and has a content-dependent normalization such that a sum of elements in the filter kernel equals a substantially fixed value.