Back to results
US 8,581,753 B2Granted
Lossless coding technique for CABAC in HEVC
AssigneeSharp Laboratories of America, Inc.
InventorsSeung-Hwan Kim, Christopher Andrew Segall
Life of this patent
Estimating the term…
Abstract
A system utilizing a lossless coding technique for CABAC in HEVC is described. The system includes a first and second electronic device. The first electronic device encodes according to a lossless coding technique for CABAC in HEVC. The second electronic device decodes according to a lossless coding technique for CABAC in HEVC.
Claims20
An apparatus, comprising:
- an electronic device configured to:
- obtain a block representing a residual sample for lossless encoding;
- generate a significance map, wherein the generating includes populating a significance map field that corresponds to the last scanning position of the block with a value corresponding to a level of the last scanning position of the block;
- generate a sequence of syntax elements including the significance map having the value;
- and
- transmit a bitstream representing the generated sequence of syntax elements to a decoder.
3 more elements
An apparatus, comprising:
- an electronic device configured to:
- obtain a block representing a residual sample for lossless encoding;
- generate a sequence of syntax elements to represent the block;
- perform adaptive binarization using a plurality of binarization tables, wherein values of an Absolute −3 portion of the sequence of syntax elements are used as input values for the plurality of binarization tables;
- encode a result of the adaptive binarization;
- and
- transmit the encoding to a decoder.
4 more elements
A method, comprising:
- filtering the context models of Context Adaptive Binary Arithmetic Coding (CABAC) based on relationship to frequency component;
- obtaining a bitstream;
- recovering a binary symbol from the bitstream;
- decoding the binary symbol using the filtered context models;
- and
- recovering video data using a result of the decoding.
3 more elements
Dependencies are inferred from the claim text.