System and method for open loop spatial prediction in a video encoder
AssigneeBroadcom Corporation
InventorDouglas Chin
Life of this patent
Estimating the term…
Abstract
Presented herein are systems, methods, and apparatus for real-time high definition television encoding. In one embodiment, there is a method for encoding video data. The method comprises downscaling at least one original reference picture; classifying the content of the macroblocks of the picture; and selecting a spatial prediction mode for each of the macroblocks of the picture based on the content classification. If it is determined that a macroblock is to be coded using spatial encoding, the macroblock is coded using the associated prediction mode.
Claims20
A method for encoding video data, the method comprising:
- classifying at least one macroblock of at least one original reference picture with a circuit;
- estimating an amount of data for encoding the at least one macroblock, in parallel, based on the macroblock classification;
- generating a target rate for encoding the at least one macroblock based on the respective estimated amount of data for the at least one macroblock;
- and
- selecting, based on the target rate or the macroblock classification, at least one quantization parameter for the at least one macroblock of the at least one original reference picture.
A system for encoding video data, the system comprising:
- a first engine for classifying at least one macroblock of at least one original reference picture;
- a second engine for estimating an amount of data for encoding the at least one macroblock, in parallel, based on the macroblock classification, and for generating a target rate for encoding the at least one macroblock based on the respective estimated amount of data for the at least one macroblock;
- and
- a quantizer for quantizing the macroblock with a quantization parameter, wherein the quantization parameters is determined at least in part based on the classification of the at least one macroblock.
A method for encoding video data, the method comprising:
- classifying at least one macroblock of at least one original reference picture, wherein classification of the at least one macro block further comprises indicating an amount of detail in the at least one macroblock, with a circuit;
- estimating an amount of data for encoding the at least one macroblock, in parallel, based on the macroblock classification;
- selecting, based on the macroblock classification, at least one quantization parameter for the at least one macroblock of the at least one original reference picture;
- and
- selecting an encoding technique for the at least one macroblock based on the amount of detail in the at least one macroblock, wherein the encoding technique is one of spatial prediction or motion estimation.
Dependencies are inferred from the claim text.