Variable length coding and decoding methods and devices for grouped pixels

Patent metadata

Field	Value
Publication number	US-10110931-B2
Application number	US-201615140085-A
Country	US
Kind code	B2
Filing date	Apr 27, 2016
Priority date	Apr 27, 2016
Publication date	Oct 23, 2018
Grant date	Oct 23, 2018

Abstract

A variable length coder is disclosed. The variable length coder comprises a size determining unit and a first residual coder. The size determining unit determines a maximum size based on sizes of quantized residuals in a current group and determines whether to enable the first residual coder according to the maximum size. When the maximum size is equal to 1, the first residual coder is enabled to encode the quantized residuals as one symbol to generate a first encoded suffix according to a variable length coding scheme.

First claim

What is claimed is:

1. A variable length coder, comprising:
a size determining unit for determining a maximum size based on sizes of quantized residuals in a current group;
the first residual coder, enabled when the maximum size is one, that encodes the quantized residuals of the current group as one symbol to generate a first encoded suffix according to a variable length coding scheme; and
a second residual coder that is enabled when the maximum size is two, wherein if the current group belongs to a pre-defined set of common combinations, the second residual coder encodes the quantized residuals of the current group as one symbol to generate a second encoded suffix according to the variable length coding scheme, otherwise, the second residual coder sequentially encodes each quantized residual of the current group using two bits to generate the second encoded suffix;
wherein the size determining unit determines whether to enable one of the first residual coder and the second residual coder according to the maximum size.

2. The variable length coder according to claim 1 , further comprising:
a grouping unit coupled to the size determining unit for dividing a plurality of quantized residuals associated with a block of pixels into a plurality of groups.

3. The variable length coder according to claim 1 , further comprising:
a third residual coder, enabled when the maximum size is three, that sequentially encodes each quantized residual of the current group to generate a third encoded suffix according to the variable length coding scheme;

wherein the size determining unit further determines whether to enable one of the first residual coder, the second residual coder and the third residual coder according to the maximum size.

4. The variable length coder according to claim 3 , further comprising:
a fourth residual coder, enabled when the maximum size is greater than three, that sequentially encodes each quantized residual of the current group using the maximum size of bits to generate a fourth encoded suffix;

wherein the size determining unit further enables one of the first residual coder, the second residual coder, the third residual coder and the fourth residual coder according to the maximum size.

5. The variable length coder according to claim 4 , further comprising:
a size prediction and encoding unit coupled to the size determining unit for receiving the maximum size and the sizes of the quantized residuals of the current group, determining a predicted size for the current group according to the sizes of the quantized residuals of a previous group, and encoding a difference between the maximum size and the predicted size to generate an encoded prefix; and
a combining unit for combining the encoded prefix and one of the first encoded suffix, the second encoded suffix, the third encode suffix and the fourth encoded suffix to generate an encoded bit stream.

6. The variable length coder according to claim 4 , further comprising:
a size encoder coupled to the size determining unit for receiving the maximum size of the current group, and encoding the maximum size to generate an encoded prefix; and
a combining unit for combining the encoded prefix and one of the first encoded suffix, the second encoded suffix, the third encoded suffix and the fourth encoded suffix to generate an encoded bit stream.

7. The variable length coder according to claim 1 , wherein the variable length coding scheme is selected from the group comprising Huffman coding, Exponential-Golomb coding, Rice coding and arithmetic coding.

8. A variable length coding method applied in a spatial prediction coding system, comprising:
determining a maximum size based on sizes of quantized residuals in a current group;
when the maximum size is one, encoding the quantized residuals of the current group as one symbol according to a variable length coding scheme to obtain a first encoded suffix; and
when the maximum size is two, encoding the quantized residuals of the current group as one symbol according to the variable length coding scheme to obtain a second encoded suffix if the current group belongs to a pre-defined set of common combinations, otherwise, sequentially encoding each quantized residual of the current group using two bits to obtain the second encoded suffix.

9. The method according to claim 8 , further comprising:
dividing a plurality of quantized residuals associated with a block of pixels into a plurality of groups according to a group volume.

10. The method according to claim 8 , further comprising:
when the maximum size is three, sequentially encoding each quantized residual of the current group according to the variable length coding scheme to obtain a third encoded suffix.

11. The method according to claim 10 , further comprising:
when the maximum size is greater than three, sequentially encodes each quantized residual of the current group using the maximum size of bits to obtain a fourth encoded suffix.

12. The method according to claim 11 , further comprising:
determining a predicted size for the current group according to the sizes of the quantized residuals of a previous group;
encoding a difference between the maximum size and the predicted size to generate an encoded prefix; and
combining the encoded prefix and one of the first encoded suffix, the second encoded suffix, the third encoded suffix and the fourth encoded suffix to obtain an encoded bit stream.

13. The method according to claim 11 , further comprising:
encoding the maximum size to obtain an encoded prefix; and
combining the encoded prefix and one of the first encoded suffix, the second encoded suffix, the third encoded suffix and the fourth encoded suffix to obtain an encoded bit stream.

14. The method according to claim 8 , wherein the variable length coding scheme is selected from the group comprising Huffman coding, Exponential-Golomb coding, Rice coding and arithmetic coding.

15. A variable length decoder, comprising:
a size decoding unit for receiving an encoded bit stream comprising an encoded prefix and an encoded suffix, and decoding the encoded prefix into a reconstructed maximum size of a current group to enable one of a first residual decoder, a second residual decoder and a third residual decoder;
the first residual coder, enabled when the reconstructed maximum size is one, that decodes the encoded suffix as one codeword into N decoded quantized residuals of the current group according to a variable length decoding scheme;
the second residual decoder that is enabled when the reconstructed maximum size is two, wherein the second residual decoder decodes the encoded suffix as one codeword into N decoded quantized residuals of the current group according to the variable length decoding scheme, and then sequentially decodes each two-bit code in the encoded suffix into a decimal value if the variable length decoding scheme fails to decode the encoded suffix; and
the third residual decoder, enabled when the reconstructed maximum size is three, that sequentially decodes each codeword in the encoded suffix according to the variable length decoding scheme;
wherein N>=2.

16. The variable length decoder according to claim 15 , further comprising:
a fourth residual decoder, enabled when the reconstructed maximum size is greater than three, that sequentially decodes each M-bit codeword in the encoded suffix;

wherein the size decoding unit further enables one of the first residual decoder, the second residual decoder, the third residual decoder and the fourth residual decoder according to the r

Assignees

• Aspeed Technology Inc

Inventors

• Lu Chung-Yen

Classifications

• H04N19/182

  the unit being a pixel · CPC title

• H04N19/18

  the unit being a set of transform coefficients · CPC title

• H04N19/13

  Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC] · CPC title

• H04N19/147

  according to rate distortion criteria (rate-distortion as a criterion for motion estimation H04N19/567) · CPC title

• H04N19/91 Primary

  Entropy coding, e.g. variable length coding [VLC] or arithmetic coding · CPC title