Adaptive variable block transform system, medium, and method

• US 20070025631A1
• Filed: 07/20/2006
• Published: 02/01/2007
• Est. Priority Date: 07/21/2005
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Birinov, Dmitri, Kim, Hyun, Cho, Daesung, Kim, Wooshik
• Granted Patent: US 7,925,107 B2
• Time in Patent Office: Days
• US Class Current: 382/248
• CPC Class Codes: H04N 19/119 Adaptive subdivision aspects, e.g. subdivision of a picture into rectangular or non-rectangular coding blocks
• CPC Class Codes: H04N 19/122 Selection of transform size, e.g. 8x8 or 2x4x8 DCT; Selection of sub-band transforms of varying structure or type
• CPC Class Codes: H04N 19/186 the unit being a colour or a chrominance component
• CPC Class Codes: H04N 19/61 in combination with predictive coding

Abstract

A variable block transform system, medium, and a system, medium, and method for video encoding/decoding using the variable block transform. The system includes a transform block size determination unit that determines the transform block size for components of an input video based on a format, defining the components, of the input video or characteristics of the components, and a transform coding unit that performs transform coding on each of the color components of the input video according to the determined transform block sizes and outputs transform coefficients.

Claims

1. A system for variable block transform, comprising:

a transform block size determination unit to designate at least one transform block size for components of an input video, based on a format of the input video, defining the components, and/or respective characteristics of the components; and

a transform coding unit to perform transform coding on each of the components according to the designated at least one transform block size.

2. The system of claim 1, wherein the transform coding unit further outputs transform coefficients based upon the performed transform coding.

3. The system of claim 1, wherein the transform block size determination unit designates the at least one transform block size for each of the components according to a predetermined rule.

4. The system of claim 3, wherein the predetermined rule includes designating a same transform block size to all components in a group, wherein components are grouped according to related characteristics, and with there being at least two groups of components.

5. The system of claim 1, wherein the transform block size determination unit designates a same transform block size for a Cb component and a Cr component when the format is a YCbCr color format.

6. The system of claim 1, wherein the transform block size determination unit designates a same transform block size for an R component, a G component, and a B component when the format is an RGB color format.

7. The system of claim 1, wherein the transform block size determination unit designates a same transform block size for an R component and a B component when the format is an RGB color format.

8. The system of claim 1, wherein transform block sizes are selectively designated to be one of 4×

4 and 8×

8.

9. A method for variable block transform, comprising:

designating at least one transform block size for components of an input video, based on a format of the input video, defining the components, and/or respective characteristics of the components; and

performing transform coding on each of the components according to the designated at least one transform block size.

10. The method of claim 9, further comprising generating transform coefficients based upon the performed transform coding.

11. The method of claim 9, wherein, in the designating of the at least one transform block size, a same transform block size is designated for a Cb component and a Cr component when the format is an YCbCr color format.

12. The method of claim 9, wherein, in the designating of the at least one transform block size, a same transform block size is designated for an R component, a G component, and a B component when the format is an RGB color format.

45. A medium comprising computer readable code to implement the method of claim 9.

13. A video encoder, comprising:

a variable block transform unit to perform transform coding on components of an input video, based on a format of the input video, defining the components, and/or respective characteristics of the components of the input video, and to output corresponding transform coefficients; and

a quantization unit to quantize the transform coefficients.

14. The video encoder of claim 13, wherein the variable block transform unit comprises:

a transform block size determination unit to designate at least one transform block size based on the format and/or the corresponding characteristics of the components; and

a transform coding unit to perform transform coding on each of the components based on the designated at least one transform block size.

15. The video encoder of claim 14, further comprising an entropy-coding unit to perform entropy-coding on the quantized coefficients and to output the entropy-coded coefficients as a bitstream.

16. The video encoder of claim 15, wherein the entropy-coding unit performs entropy-coding on information about the designated at least one transform block size.

17. The video encoder of claim 16, wherein the transform block size determination unit designates the designated at least one transform block size for each of the components according to a predetermined rule, and

the entropy-coding unit performs the entropy-coding on the information about the designated at least one transform block size, and outputs the entropy-coded information as a bitstream.

18. The video encoder of claim 17, wherein the predetermined rule includes designating a same transform block sizes to all components in a group, wherein components are grouped according to related characteristics, and with there being at least two groups of components, and

the entropy-coding unit performs the entropy-coding on the information about the designated at least one transform block size, as designated for each group.

19. The video encoder of claim 16, wherein the transform block size determination unit designates a same transform block size for a Cb component and a Cr component when the format is an YCbCr color format, and

the entropy-coding unit performs the entropy-coding on only one of information about the same transform block size of the Cb component and information about the same transform block size of the Cr component.

20. The video encoder of claim 16, wherein the transform block size determination unit designates a same transform block size for an R component, a G component, and a B component when the format is an RGB color format, and

the entropy-coding unit performs the entropy-coding on only one of information about the same transform block size of the R component, information about the same transform block size of the G component, and information about the same transform block size of the B component.

21. The video encoder of claim 14, wherein the transform coding unit collects DC coefficients for each of color components from the transform coefficients, performs Hadamard transform on the collected DC coefficients, and outputs the Hadamard-transformed transform coefficients.

22. The video encoder of claim 21, wherein, when the at least one designated transform block size is 8×

8, the transform coding unit collects four DC coefficients for each of the components from the transform coefficients in four 8×

8 blocks, performs 2×

2 Hadamard transform on the collected DC coefficients, and outputs the Hadamard-transformed transform coefficients.

23. A video encoding method, comprising:

performing transform coding on components of an input video based on a format of the input video, defining the components, and/or respective characteristics of the components, and generating corresponding transform coefficients; and

quantizing the transform coefficients for generating an encoded video bitstream.

24. The video encoding method of claim 23, wherein the performing of the transform coding comprises:

designating at least one transform block size based on the format and/or the respective characteristics of the components; and

performing transform coding on each of the components of the input video based on the designated at least one transform block size.

25. The video encoding method of claim 24, further comprising performing entropy-coding on the quantized coefficients and outputting the entropy-coded coefficients as a bitstream.

26. The video encoding method of claim 25, wherein, in the performing of the entropy-coding, the entropy-coding is performed on information about the designated at least one transform block size

27. The video encoding method of claim 26, wherein,

in the designating of the at least one transform block size, a same transform block size is designated for a Cb component and a Cr component when the format is an YCbCr color format, and

in the performing of the entropy-coding, the entropy-coding is performed on only one of information about the same transform block size of the Cb component and information about the same transform block size of the Cr component.

28. The video encoding method of claim 26, wherein,

in the designating of the at least one transform block size, a same transform block size is designated for an R component, a G component, and a B component when the format is an RGB color format, and

in the performing of the entropy-coding, the entropy-coding is performed on only one of information about the same transform block size of the R component, information about the same transform block size of the G component, and information about the same transform block size of the B component.

46. A medium comprising computer readable code to implement the method of claim 23.

29. A video decoder for decoding encoded input video data, comprising:

a variable block inverse transform unit to perform variable block inverse transform on inversely quantized data related to components of the encoded input video data based on a format of the encoded input video data, defining the components, or respective characteristics of the components using predetermined corresponding transform block size information provided for and/or with the encoded input video data.

30. The video decoder of claim 29, further comprising an inverse quantization unit to inversely quantize the encoded input video data.

31. The video decoder of claim 29, wherein the variable block inverse transform unit comprises:

a transform block size determination unit determine at least one previously designated transform block size for the components based on the format and/or the respective characteristics of each of the components, using the provided predetermined corresponding transform block size information; and

an inverse transform coding unit to perform inverse transform coding on each of the components of the inversely quantized data based on the determined at least one transform block size.

32. The video decoder of claim 31, further comprising an entropy-decoding unit to perform entropy-decoding an input bitstream and to output the encoded input video data.

33. The video decoder of claim 32, wherein the entropy-decoding unit performs entropy-decoding of the input bitstream to decode the provided predetermined corresponding transform block size information.

34. The video decoder of claim 31, wherein the transform block size determination unit determines the previously designated transform block size for each of the components based on a predetermined rule using the provided predetermined corresponding transform block size information.

35. The video decoder of claim 34, wherein

the predetermined rule includes having a same transform block size designated to all components in a group, wherein components are grouped according to related characteristics, and with there being at least two groups of components, and

the transform block size determination unit extracts transform block size of each group using the provided predetermined corresponding transform block size information and correspondingly determines the previously designated transform block size for each of the components.

36. The video decoder of claim 31, wherein the transform block size determination unit determines a same transform block size was previously designated for a Cb component and a Cr component when the format is an YCbCr color format, using the provided predetermined corresponding transform block size information.

37. The video decoder of claim 31, wherein the transform block size determination unit determines a same transform block size was previously designated for an R component, a G component, and a B component when the format is an RGB color format, using the provided predetermined corresponding transform block size information.

38. The video decoder of claim 31, wherein the inverse transform coding unit collects DC coefficients for each of the components of the inversely quantized data and performs inverse transform coding on data obtained through an inverse Hadamard transform of the collected DC coefficients and AC coefficients.

39. A video decoding method, comprising:

performing transform coding on components of an encoded input video based on a format of the encoded input video, defining the components, and/or respective characteristics of the components of the encoded input video and generating corresponding transform coefficients; and

quantizing the transform coefficients for decoding the encoded input video.

40. The video decoding method of claim 39, wherein the quantizing of the transform coefficients comprises:

determining at least one previously designated transform block size for the components based on the format and/or each of the respective characteristics of the components using predetermined corresponding transform block size information provided for and/or with the encoded input video data; and

performing inverse transform coding on each of the components of the inversely quantized data according to the determined at least one transform block size.

41. The video decoding method of claim 40, further comprising performing entropy-decoding on an input bitstream to obtain the encoded input video data.

42. The video decoding method of claim 41, wherein, in the performing of the entropy-decoding, the entropy-decoding is performed on the input bitstream to obtain the predetermined corresponding transform block size information.

43. The video decoding method of claim 40, wherein, in the determining of the at least one transform block size, a same transform block size is determined for a Cb component and a Cr component when the format is an YCbCr color format, using the predetermined corresponding transform block size information.

44. The video decoding method of claim 40, wherein, in the determining of the at least one transform block size, a same transform block size is determined for an R component, a G component, and a B component when the format is an RGB color format, using the predetermined corresponding transform block size information.

47. A medium comprising computer readable code to implement the method of claim 39.