Back to results
US 10,321,163 B2Granted
High definition VP8 decoder
AssigneeTexas Instruments Incorporated
InventorsMihir Narendra Mody, Chaitanya S Ghone, Joseph Meehan
Life of this patent
Estimating the term…
Abstract
A VP8 video decoder is implemented by partitioning the required functions across multiple sub systems, with an optimal mapping to existing functional blocks. Key optimizations include the reuse of hardware designed for prior generation VP6 and VP7 decoders. In order to reduce implementation complexity, cost and power consumption, a non-exact, approximate deblocking loop filter is implemented.
Claims13
A video decoding system comprising:
- one or more processors configured to:
- perform entropy decoding in a frame level pipelined manner on a bitstream representing encoded video data to produce a first output;
- perform at least one of an inverse quantization operation or an inverse frequency transform operation in a row level pipelined manner on the first output to produce a second output;
- perform deblocking filtering on an input that includes the second output in a row level pipelined manner using a non-exact approximate loop filter to produce a third output;
- apply a fading compensation operation in a row level pipelined manner to the third output to produce a fourth output;
- apply a motion compensation operation in a row level pipelined manner to the fourth output to produce a fifth output;
- combine the fifth output with the second output by a feedback path;
- and
- output a decoded video output based on the third output.
6 more elements
An electronic device comprising:
- a display device;
- and
- one or more processors configured to:
- perform entropy decoding in a frame level pipelined manner on a bitstream representing encoded video data to produce a first output;
- perform at least one of an inverse quantization operation or an inverse frequency transform operation in a row level pipelined manner on the first output to produce a second output;
- perform deblocking filtering on an input that includes the second output in a row level pipelined manner using a non-exact approximate loop filter to produce a third output;
- apply a fading compensation operation in a row level pipelined manner to the third output to produce a fourth output;
- apply a motion compensation operation in a row level pipelined manner to the fourth output to produce a fifth output;
- combine the fifth output with the second output by a feedback path;
- produce a decoded video output based on the third input;
- and
- cause the decoded video output to be displayed on the display device.
9 more elements
A video decoding system comprising:
- one or more processors configured to:
- perform entropy decoding in a frame level pipelined manner on a bitstream representing encoded video data to produce a first output;
- perform at least one of an inverse quantization operation or an inverse frequency transform operation on the first output to produce a second output;
- perform deblocking filtering on an input that includes the second output using a non-exact approximate loop filter to produce a third output;
- apply a fading compensation operation in a row level pipelined manner to the third output to produce a fourth output;
- apply a motion compensation operation to the fourth output to produce a fifth output;
- combine the fifth output with the second output by a feedback path;
- and
- output a decoded video output based on the third output;
- wherein the one or more processors comprises a first processor that is an embedded digital processor and a second processor that is a digital signal processor, performing entropy decoding is performed by the first processor, performing at least one of the inverse quantization operation or the inverse frequency transform operation is performed by the second processor;
- and
- wherein each of performing at least one of the inverse quantization operation or the inverse frequency transform operation, performing deblocking filtering, and applying the motion compensation operation are performed in a row level pipelined manner.
9 more elements
Dependencies are inferred from the claim text.