General Instrument Corporation

Google Technology Holdings LLC

Yendo Hu

Jing Yang Chen

Robert S Nemiroff

John Shumate

Shanhua Xue

A video coding system includes a decoder configured to decode an encoded reference picture, a primary memory, and a secondary memory configured to store the decoded reference picture. The video coding system also includes a logic device configured to extract motion vectors for generating a predicted picture of the reference picture and to store the extracted motion vectors in memory bins. The reference picture includes a plurality of predefined regions and each memory bin stores motion vectors for a particular region of the plurality of predefined regions. For each of the plurality of predefined regions of the reference picture, the logic device is configured to use the memory bin corresponding to a reference region to read the reference region from the secondary memory and write the reference region into the primary memory. The written reference region and the motion vectors of the corresponding memory bin are used to spatially translate the written reference region into the predicted picture.

Motion compensation in video coding

A method for high/low usage is provided. The method receives a macroblock data structure and a syntax element at a digital signal processing engine. Further, the method classifies the syntax element as high use or low use. In addition, the method sends the syntax element from the digital signal processing engine to a logic unit, distinct from the digital processing engine, for binarization if the syntax element is high use.

Method and apparatus for context address generation for motion vectors and coefficients

A method stores, in a first memory location in a context weight update engine in an arithmetic encoder, a context weight value. The method also stores, in a second memory location in the context weight update engine in the arithmetic encoder, a context weight adjustment value. Further, the method reads, in a first clock cycle, the context weight value and a first binarization value from the first memory location. In addition, the method writes, in a second clock cycle, the context weight adjustment value and a second binarization value into the first memory location. The second binarization value is distinct from the first binarization value.

Method and system for providing single cycle context weight update leveraging context address look ahead

A method and system are provided for code normalization and byte construction. A plurality of subsets of bits is extracted from a first input. Each of the subsets of bits has a bit width equaling a number of leading zeros from a second input variable. Further, a consecutive sequence of the plurality of subsets is stored in a memory. In addition, the consecutive sequence of the plurality of subsets is read from the memory if a third input release flag is established.

Growing community of inventors

La Jolla, CA, United States of America

Yendo Hu