Buffering apparatus for buffering multi-partition video/image bitstream and related method thereof

Patent metadata

Field	Value
Publication number	US-9497466-B2
Application number	US-201113304372-A
Country	US
Kind code	B2
Filing date	Nov 24, 2011
Priority date	Jan 17, 2011
Publication date	Nov 15, 2016
Grant date	Nov 15, 2016

Abstract

An exemplary buffering apparatus for buffering a multi-partition video/image bitstream which transmits a plurality of compressed frames each having a plurality of partitions includes a first bitstream buffer and a second bitstream buffer. The first bitstream buffer is arranged to buffer data of a first partition of the partitions of a specific compressed frame. The second bitstream buffer is arranged to buffer data of a second partition of the partitions of the specific compressed frame.

First claim

What is claimed is:

1. A buffering apparatus for buffering a multi-partition video/image bitstream which transmits a plurality of compressed frames, each compressed frame comprising a plurality of macroblocks and being divided into a plurality of partitions, the buffering apparatus comprising:
a first bitstream buffer, arranged to buffer data of a first partition of the partitions of a specific compressed frame;
a second bitstream buffer, arranged to buffer data of a second partition of the partitions of the specific compressed frame;
a barrel shifter, coupled to the first bitstream buffer and the second bitstream buffer, for parsing data of the specific compressed frame in order to sequentially decode macroblocks of the bitstream;
a multiplexer, coupled to the first bitstream buffer, the second bitstream buffer and the barrel shifter, for selectively outputting data of the first partition or data of the second partition; and
a Direct Memory Access (DMA) controller, coupled between the multiplexer and the barrel shifter, for transmitting the data of the first partition or the data of the second partition to the barrel shifter in a DMA manner;
wherein the partitions of the specific compressed frame are transmitted individually to the barrel shifter, the first partition and the second partition are two of remaining partitions that are transmitted after a leading partition, and data of a leading partition includes header information applied to the whole specific compressed frame and prediction information for each macroblock in the specific compressed frame.

2. The buffering apparatus of claim 1 , wherein the first and second bitstream buffers are ring buffers dedicated to respectively buffering data of the first partition and data of the second partition.

3. The buffering apparatus of claim 1 , wherein the data of the first partition includes transform coefficients; and the data of the second partition includes transform coefficients.

4. The buffering apparatus of claim 1 , wherein the multi-partition video/image bitstream is a VP8/WebP bitstream.

5. The buffering apparatus of claim 1 , wherein the buffering apparatus further comprises:
a prefetch circuit, arranged to prefetch data and store the prefetched data into the second bitstream buffer while the first bitstream buffer is buffering the data of the first partition.

6. The buffering apparatus of claim 1 , further comprising:
a prefetch circuit, comprising:
a storage unit; and
a prefetch unit, arranged to prefetch data from the second bitstream buffer and store the prefetched data into the storage unit while a decoding apparatus is decoding the first partition, wherein the prefetched data stored in the prefetch unit is read by the decoding apparatus when the decoding apparatus starts decoding the second partition.

7. A method for buffering a multi-partition video/image bitstream which transmits a plurality of compressed frames, each compressed frame comprising a plurality of macroblocks and being divided into a plurality of partitions, the method comprising:
utilizing a first bitstream buffer to buffer data of a first partition of the partitions of a specific compressed frame;
utilizing a second bitstream buffer to buffer data of a second partition of the partitions of the specific compressed frame;
utilizing a barrel shifter coupled to the first bitstream buffer and the second bitstream buffer to parse data of the specific compressed frame for sequential decoding of macroblocks of the bitstream;
utilizing a multiplexer coupled to the first bitstream buffer, the second bitstream buffer and the barrel shifter to selectively output data of the first partition or data of the second partition; and
utilizing a Direct Memory Access (DMA) controller, coupled between the multiplexer and the barrel shifter, to transmit the data of the first partition or the data of the second partition to the barrel shifter;
wherein the partitions of the specific compressed frame are transmitted individually to the barrel shifter, the first partition and the second partition are two of remaining partitions that are transmitted after a leading partition, and data of a leading partition includes header information applied to the whole specific compressed frame and prediction information for each macroblock in the specific compressed frame.

8. The method of claim 7 , wherein the step of utilizing the first bitstream buffer comprises: utilizing a ring buffer dedicated to buffering data of the first partition as the first bitstream buffer; and the step of utilizing the second bitstream buffer comprises: utilizing a ring buffer dedicated to buffering data of the second partition as the second bitstream buffer.

9. The method of claim 7 , wherein the data of the first partition includes transform coefficients; and the data of the second partition includes transform coefficients.

10. The method of claim 7 , wherein the multi-partition video/image bitstream is a VP8/WebP bitstream.

11. The method of claim 7 , wherein the method further comprises:
prefetching data and storing the prefetched data into the second bitstream buffer while the first bitstream buffer is buffering the data of the first partition.

12. The method of claim 7 , further comprising:
prefetching data from the second bitstream buffer and buffering the prefetched data while decoding of the first partition is being executed, wherein the buffered prefetched data is read when decoding of the second partition is started.

Assignees

• Cheng Chia-Yun
• Wang Sheng-Jen
• Tseng Yun-Feng
• Cui Jun
• Mediatek Inc

Inventors

• Cheng Chia-Yun
• Wang Sheng-Jen
• Tseng Yun-Feng
• Cui Jun

Classifications

• H04N19/436

  using parallelised computational arrangements · CPC title

• G06F13/385

  for adaptation of a particular data processing system to different peripheral devices · CPC title

• H04N19/176

  the region being a block, e.g. a macroblock · CPC title

• H04N19/44

  Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder · CPC title

• G06F5/10

  having a sequence of storage locations each being individually accessible for both enqueue and dequeue operations, e.g. using random access memory {(G06F5/065 takes precedence)} · CPC title