Unified fractional search and motion compensation architecture across multiple video standards

摘要

Methods and systems for performing at least one of video encoding and video decoding are disclosed. In one implementation, the system includes a controller configured to determine a video standard associated with a portion of the video data, each portion of the video data associated with one of a plurality of video standards. The controller is further configured to provide a set of the filter parameters which are associated with a video standard to be used for at least one of the video encoding and decoding and at least one filter configured to filter at least one reference pixel received from the reference pixel memory based, at least in part, on the provided set of filter parameters.

主权项

1. A system for processing video data, the system comprising:
a controller configured to:
determine a video standard associated with a portion of the video data, each portion of the video data associated with one of a plurality of video standards, the plurality of video standards comprising at least one of VP6, VP8, and HEVC;provide a set of filter parameters for the portion of the video data based on the determined video standard; a plurality of filters configured to filter at least one reference pixel included in the portion of the video data based, at least in part, on the provided set of filter parameters, the plurality of filters comprising a first filter, a second filter, and a third filter; and
a half pixel buffer,wherein the first filter is configured to filter the at least one reference pixel based on at least one of the provided filter parameters, the second filter is configured to filter the output of the first filter based on at least one of the provided filter parameters, the half pixel buffer is configured to buffer the output of the second filter, and the third filter is configured to filter the output of the half pixel buffer,wherein the filters are configurably pipelined according to the determined video standard,
wherein, if the determined video standard is of a first type, filtering the at least one reference pixel comprises:
horizontally filtering, via the first filter, the at least one reference pixel; andvertically filtering, via the second filter, the output of the first filter,wherein, if the determined video standard is of a second type, filtering the at least one reference pixel comprises:
vertically filtering, via the first filter, the at least one reference pixel; andhorizontally filtering, via the second filter, the output of the first filter,wherein, if the determined video standard is of a third type, filtering the at least one reference pixel comprises:
horizontally and vertically filtering, via the first filter, the at least one reference pixel;vertically filtering, via the second filter, the output of the first filter; andbilinear filtering, via the third filter, the output of the half pixel buffer,wherein processing the video data comprises at least one of encoding the video data and decoding the video data, andwherein the controller is configured to determine the video standard based, at least in part, on video standard identification information contained in a fractional search engine (FSE) software register.