Back to results
US 9,426,498 B2Granted
Real-time encoding system of multiple spatially scaled video based on shared video coding information
AssigneeBroadcom Corporation
InventorLei Zhang
Life of this patent
Estimating the term…
Abstract
In one embodiment, a method comprising receiving at a single encoding engine an input video stream having one or more pictures of a first size; and generating by the single encoding engine, in parallel, plural encoded streams, a first of the encoded streams comprising one or more pictures of the first size and a second of the encoded streams comprising one or more pictures of a second size that is smaller than the first size, the encoding of the second stream based on sharing video coding information used in encoding the first encoded stream.
Claims17
A method, comprising:
- receiving at a single encoding engine an input video stream having one or more pictures of a first resolution;
- and
- generating by the single encoding engine, in parallel, plural encoded streams, a first of the encoded streams comprising one or more pictures of the first resolution and a second of the encoded streams comprising one or more pictures of a second resolution that is smaller than the first resolution, the encoding of the second stream based on sharing video coding information used in encoding the first encoded stream, wherein a plurality of first blocks of the first encoded stream are downscaled by a scaling factor to derive a plurality of second blocks of the second encoded stream;
- mapping a plurality of first motion vectors of the first blocks to a plurality of second motion vectors of the second blocks, the second motion vectors each corresponding to one of the first motion vectors adjusted by the scaling factor;
- and
- partitioning at least one of a macroblock or a coding unit of the second encoded stream into one or more partitions, the at least one of the macroblock or the coding unit comprising a plurality of the second blocks, the at least one of the macroblock or the coding unit partitioned by:
- identifying a plurality of neighboring second blocks within the at least one of the macroblock or the coding unit having equal second motion vectors;
- and
- combining the neighboring second blocks having the equal second motion vectors within a partition of the one or more partitions.
6 more elements
A system comprising:
- a single encoding engine configured to:
- receive an input video stream having one or more pictures of a first resolution;
- generate in parallel plural encoded streams, the a first encoded stream of the encoded streams comprising one or more pictures of the first resolution and a second of the encoded streams comprising one or more pictures of a second resolution that is smaller than the first resolution, the encoding of the second stream based on sharing video coding information used in encoding the first encoded stream
- map a plurality of first motion vectors of the first blocks to a plurality of second motion vectors of the second blocks, the second motion vectors each corresponding to one of the first motion vectors adjusted by the scaling factor;
- and
- partition at least one of a macroblock or a coding unit of the second encoded stream into one or more partitions, the at least one of the macroblock or the coding unit comprising a plurality of the second blocks, the single encoding engine configured to partition at least one of the macroblock or the coding unit by:
- identifying a plurality of neighboring second blocks within the at least one of the macroblock or the coding unit having a same second motion vector;
- and
- combining the neighboring second blocks having the same second motion vector within a partition of the one or more partitions.
6 more elements
Dependencies are inferred from the claim text.