Method for sub-pixel value interpolation

• US 8,036,273 B2
• Filed: 08/15/2007
• Issued: 10/11/2011
• Est. Priority Date: 09/17/2001
• Status: Expired due to Fees

• Current Assignee: Nokia Technologies Oy (Nokia Corporation)
• Original Assignee: Nokia Corporation
• Inventors: Karczewicz, Marta, Hallapuro, Antti
• Primary Examiner(s): An; Shawn
• Application Number:
  US11/839,205
• Publication Number:
  US 20080069203A1
• Time in Patent Office: 1,518 Days
• Field of Search: 375/240.17, 375/240.16, 375/240.12, 375/240.25, 375/240.26, 382/238, 382/236, 382/233, 382/235
• US Class Current: 375/240.17
• CPC Class Codes: G06T 3/4007 based on interpolation, e.g. bilinear interpolation image demosaicing G06T3/4015; edge-driven or edge-based scaling G06T3/403
• CPC Class Codes: H04N 19/523 with sub-pixel accuracy
• CPC Class Codes: H04N 19/80 Details of filtering operations specially adapted for video compression, e.g. for pixel interpolation H04N19/635, H04N19/86 take precedence

Abstract

A method of interpolation in video coding in which an image comprising pixels arranged in rows and columns and represented by values having a specified dynamic range, the pixels in the rows residing at unit horizontal locations and the pixels in the columns residing at unit vertical locations, is interpolated to generate values for sub-pixels at fractional horizontal and vertical locations, the method comprising:

• a) when values for sub-pixels at half unit horizontal and unit vertical locations, and unit horizontal and half unit vertical locations are required, interpolating such values directly using weighted sums of pixels residing at unit horizontal and unit vertical locations;
• b) when values for sub-pixels at half unit horizontal and half unit vertical locations are required, interpolating such values directly using a weighted sum of values for sub-pixels residing at half unit horizontal and unit vertical locations calculated according to step (a); and
• c) when values for sub-pixels at quarter unit horizontal and quarter unit vertical locations are required, interpolating such values by taking the average of at least one pair of a first pair of values of a sub-pixel located at a half unit horizontal and unit vertical location, and a sub-pixel located at a unit horizontal and half unit vertical location and a second pair of values of a pixel located at a unit horizontal and unit vertical location, and a sub-pixel located at a half unit horizontal and half unit vertical location.

Claims

1. A method for sub-pixel value interpolation to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically within the rectangular bounded region using the co-ordinate notation K/2^N, L/2^N, K and L being positive integers having respective values between zero and 2^N, N being a positive integer greater than one and representing a particular degree of sub-pixel value interpolation, the method comprising causing an apparatus to:

interpolate a sub-pixel value for a sub-pixel having co-ordinates with odd values of K and L, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates ½

, ½

, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with even values of K and L, including zero, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co-ordinates ½

, ½ and

the nearest neighbouring pixel;

interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to an even value, used in the interpolation of the sub-pixels having co-ordinates with odd values of K and L, using weighted sums of the values of pixels located in rows and columns respectively; and

interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a predetermined choice of either a weighted sum of the values of sub-pixels having co-ordinates with K equal to an even value and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, or a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to an even value and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions.

2. A method according to claim 1, comprising causing the apparatus to use a first and a second weight in the weighted average chosen to interpolate the sub-pixel value for a sub-pixel having co-ordinates with odd values of K and L, wherein when the chosen weighted average is that involving the value of a nearest neighbouring pixel and the value of the sub-pixel situated at co-ordinates ½

, ½

, causing the apparatus to select the relative magnitudes of the first and second weights to be inversely proportional to the respective straight-line diagonal distances of the nearest-neighbouring pixel and the sub-pixel situated at co-ordinates ½

, ½

to the sub-pixel having co-ordinates with odd values of K and L whose value is being interpolated.

3. A method according to claim 2, comprising causing the apparatus to use first and second weights with equal values when the nearest-neighbouring pixel and sub-pixel pixel situated at co-ordinates ½

, ½

are equidistant from the sub-pixel having co-ordinates with odd values of K and L whose value is being interpolated.

4. A method according to claim 1, comprising causing the apparatus to use a first and a second weight in the weighted average chosen to interpolate the sub-pixel value for a sub-pixel having co-ordinates with odd values of K and L, wherein when the chosen weighted average is that involving the values of a pair of diagonally-opposed sub-pixels having co-ordinates with even values of K and L, causing the apparatus to select the relative magnitudes of the first and second weights to be inversely proportional to the respective straight-line diagonal distances of the sub-pixels having co-ordinates with even values of K and L to the sub-pixel having co-ordinates with odd values of K and L whose value is being interpolated.

5. A method according to claim 4, comprising causing the apparatus to use first and second weights with equal values when the sub-pixels having co-ordinates with even values of K and L are equidistant from the sub-pixel having co-ordinates with odd values of K and L whose value is being interpolated.

6. A method according to claim 1, comprising causing the apparatus to interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a weighted sum of the values of sub-pixels having co-ordinates with K equal to an even value and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions when the value of a sub-pixel having co-ordinates with K equal to an even value and L equal to an odd value is also required.

7. A method according to claim 1, comprising causing the apparatus to interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to an even value and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions when the value of a sub-pixel having co-ordinates with K equal to an odd value and L equal to an even value is also required.

8. A method according to claim 1, comprising causing the apparatus to interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an odd value O and L equal to an even value E, not including zero or 2^N, by taking an average of the value of a first sub-pixel having co-ordinates with K equal to the even value O−

1 and L equal to E and a second sub-pixel having co-ordinates with K equal to the even value O+1 and L equal to E.

9. A method according to claim 1, comprising causing the apparatus to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to 1 and L equal to zero, by taking an average of the value of a pixel having co-ordinates with K equal to zero and L equal to zero, and a sub-pixel having co-ordinates with K equal to 2 and L equal to zero.

10. A method according to claim 1, comprising causing the apparatus to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to 2^N−

1 and L equal to zero, by taking an average of the value of a pixel having co-ordinates with K equal to 2^N and L equal to zero, and a sub-pixel having co-ordinates with K equal to 2^N−

2 and L equal to zero.

11. A method according to claim 1, comprising causing the apparatus to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to 1 and L equal to 2^N, by taking an average of the value of a pixel having co-ordinates with K equal to zero and L equal to 2^N, and a sub-pixel having co-ordinates with K equal to 2 and L equal to 2^N.

12. A method according to claim 1, comprising causing the apparatus to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to 2^N−

1 and L equal to 2^N, by taking an average of the value of a pixel having co-ordinates with K equal to 2^N and L equal to 2^N, and a sub-pixel having co-ordinates with K equal to 2^N−

2 and L equal to 2^N.

13. A method according to claim 1, comprising causing the apparatus to interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an even value E, not including zero or 2^N, and L equal to an odd value O, by taking an average of the value of a first sub-pixel having co-ordinates with K equal to E and L equal to the even value O−

1 and a second sub-pixel having co-ordinates with K equal to E and L equal to the even value O+1.

14. A method according to claim 1, comprising causing the apparatus to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to zero and L equal to 1, by taking an average of the value of a pixel having co-ordinates with K equal to zero and L equal to zero, and a sub-pixel having co-ordinates with K equal to zero and L equal to 2.

15. A method according to claim 1, comprising causing the apparatus to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to 2^N and L equal to 1, by taking an average of the value of a pixel having co-ordinates with K equal to 2^N and L equal to zero, and a sub-pixel having co-ordinates with K equal to 2^N and L equal to 2.

16. A method according to claim 1, comprising causing the apparatus to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to zero and L equal to 2^N−

1, by taking an average of the value of a pixel having co-ordinates with K equal to zero and L equal to 2^N, and a sub-pixel having co-ordinates with K equal to zero and L equal to 2^N−

2.

17. A method according to claim 1, comprising causing the apparatus to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to 2^N and L equal to 2^N−

1, by taking an average of the value of a pixel having co-ordinates with K equal to 2^N and L equal to 2^N, and a sub-pixel having co-ordinates with K equal to 2^N and L equal to 2^N−

2.

18. A method according to claim 1, comprising causing the apparatus to interpolate a sub-pixel value for the sub-pixel having co-ordinates with K equal to 2^N−

1 and L equal to 2^N−

1, by taking a weighted average of the values of the four corner pixels that define the rectangular bounded region.

19. A method according to claim 1, in which N is one of the values 2, 3, and 4.

20. A method according to claim 1, wherein the pixels have a predetermined dynamic range, said predetermined dynamic range corresponding to the range of values said pixels can take.

21. A method according to claim 20, comprising interpolating sub-pixel values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and for sub-pixels having co-ordinates with K equal to zero and L equal to an even value, using weighted sums of the values of pixels located in rows and columns respectively, to produce a first intermediate value, said first intermediate value having a dynamic range equal to the predetermined dynamic range of the pixels used in said weighted sum multiplied by a value equal to the sum of the respective weights used in said weighted sum.

22. A method according to claim 21, wherein the dynamic range of said first intermediate value is defined by the number of bits used to represent the respective pixels used in said weighted sum plus the number of bits required to represent the sum of the respective weights used in said weighted sum.

23. A method according to claim 21, comprising truncating the first intermediate value by mathematically dividing the first intermediate value by a first scale factor, said first scale factor having a value equal to the sum of the respective weights used in said weighted sum, thereby forming a sub-pixel value with a dynamic range equal to the predetermined dynamic range of said pixels.

24. A method according to claim 23, comprising using a truncated first intermediate sub-pixel value when interpolating a value for a sub-pixel having co-ordinates with odd values of K and L.

25. A method according to claim 21, comprising interpolating sub-pixel values for sub-pixels having co-ordinates with even values of K and L, using a weighted sum comprising first intermediate values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and corresponding first intermediate values for sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, to produce a second intermediate value, said second intermediate value having a dynamic range equal to the dynamic range of the first intermediate values used in said weighted sum, multiplied by a value equal to the sum of the respective weights used in said weighted sum.

26. A method according to claim 25, comprising truncating the second intermediate value by mathematically dividing the second intermediate value by a second scale factor, said second scale factor having a value equal to the sum of the respective weights used in the weighted sum of claim 21 multiplied by the sum of the respective weights used in the weighted sum of claim 25, thereby forming a sub-pixel value with a dynamic range equal to the predetermined dynamic range of said pixels.

27. A method according to claim 26, comprising using a truncated second intermediate sub-pixel value when interpolating a value for a sub-pixel having co-ordinates with odd values of K and L.

28. A method according to claim 21, comprising interpolating sub-pixel values for sub-pixels having co-ordinates with even values of K and L, using a weighted sum comprising first intermediate values for sub-pixels having co-ordinates with K equal to zero and L equal to an even value and corresponding first intermediate values for sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions to produce a second intermediate value, said second intermediate value having a dynamic range equal to the dynamic range of the first intermediate values used in said weighted sum, multiplied by a value equal to the sum of the respective weights used in said weighted sum.

29. A method according to claim 28, comprising truncating the second intermediate value by mathematically dividing the second intermediate value by a second scale factor, said second scale factor having a value equal to the sum of the respective weights used in the weighted sum of claim 21 multiplied by the sum of the respective weights used in the weighted sum of claim 28, thereby forming a sub-pixel value with a dynamic range equal to the dynamic range of said pixels.

30. A method according to claim 29, comprising using a truncated second intermediate sub-pixel value when interpolating a value for a sub-pixel having co-ordinates with odd values of K and L.

31. A method for quarter resolution sub-pixel value interpolation to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically within the rectangular bounded region using the co-ordinate notation K/4, L/4, K and L being positive integers having respective values between zero and 4, the method comprising causing an apparatus to:

interpolate a sub-pixel value for a sub-pixel having co-ordinates with K and L equal to 1 or 3, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates 2/4, 2/4, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with K and L equal to zero, 2 or 4, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co-ordinates 2/4, 2/4 and the nearest neighbouring pixel;

interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to 2 and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to 2, used in the interpolation of the sub-pixels having co-ordinates with K and L equal to 1 or 3, using weighted sums of the values of pixels located in rows and columns respectively; and

interpolate a sub-pixel value for the sub-pixel having co-ordinates with both K and L equal to 2, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with K and L equal to 1 or 3, using a predetermined choice of either a weighted sum of the values of the sub-pixel having co-ordinates with K equal to 2 and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, or a weighted sum of the value of the sub-pixel having co-ordinates with K equal to zero and L equal to 2 and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions.

32. A method for eighth resolution sub-pixel value interpolation to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically with the rectangular bounded region using the co-ordinate notation K/8, L/8, K and L being positive integers having respective values between zero and 8, the method comprising causing an apparatus to:

interpolate a sub-pixel value for a sub-pixel having co-ordinates with K and L equal to 1, 3, 5 or 7, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates 4/8, 4/8, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with K and L equal to zero, 2, 4, 6 or 8, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co-ordinates 4/8, 4/8 and the nearest neighbouring pixel;

interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to 2, 4 or 6 and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to 2, 4 or 6, used in the interpolation of the sub-pixels having co-ordinates with K and L equal to 1, 3, 5 or 7, using weighted sums of the values of pixels located in rows and columns respectively; and

interpolate sub-pixel values for sub-pixels having co-ordinates with K and L equal to 2, 4 or 6, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with K and L equal to 1, 3, 5 or 7, using a predetermined choice of either a weighted sum of the values of sub-pixels having co-ordinates with K equal to 2, 4 or 6 and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions or a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to 2, 4 or 6 and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions.

33. An interpolator for sub-pixel value interpolation, the interpolator being configured to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically within the rectangular bounded region using the co-ordinate notation K/2^N, L/2^N, K and L being positive integers having respective values between zero and 2^N, N being a positive integer greater than one and representing a particular degree of sub-pixel value interpolation, the interpolator being configured to:

interpolate a sub-pixel value for a sub-pixel having co-ordinates with odd values of K and L, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates ½

, ½

, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with even values of K and L, including zero, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co-ordinates ½

, ½ and

the nearest neighbouring pixel;

interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to an even value, used in the interpolation of the sub-pixels having co-ordinates with odd values of K and L, using weighted sums of the values of pixels located in rows and columns respectively; and

interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a predetermined choice of either a weighted sum of the values of sub-pixels having co-ordinates with K equal to an even value and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, or a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to an even value and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions.

34. An interpolator according to claim 33, wherein the interpolator is configured to use a first and a second weight in the weighted average chosen to interpolate the sub-pixel value for a sub-pixel having co-ordinates with odd values of K and L, wherein when the chosen weighted average is that involving the value of a nearest neighbouring pixel and the value of the sub-pixel situated at co-ordinates ½

, ½

, the interpolator is configured to select the relative magnitudes of the first and second weights to be inversely proportional to the respective straight-line diagonal distances of the nearest-neighbouring pixel and the sub-pixel situated at co-ordinates ½

, ½

to the sub-pixel having co-ordinates with odd values of K and L whose value is being interpolated.

35. An interpolator according to claim 34, wherein the interpolator is configured to use first and second weights with equal values when the nearest-neighbouring pixel and sub-pixel situated at co-ordinates ½

, ½

are equidistant from the sub-pixel having co-ordinates with odd values of K and L whose value is being interpolated.

36. An interpolator according to claim 33, wherein the interpolator is configured to use a first and a second weight in the weighted average chosen to interpolate the sub-pixel value for a sub-pixel having co-ordinates with odd values of K and L, wherein when the chosen weighted average is that involving the values of a pair of diagonally-opposed sub-pixels having co-ordinates with even values of K and L, the interpolator is configured to select the relative magnitudes of the first and second weights to be inversely proportional to the respective straight-line diagonal distances of the sub-pixels having co-ordinates with even values of K and L to the sub-pixel having co-ordinates with odd values of K and L whose value is being interpolated.

37. An interpolator according to claim 36, wherein the interpolator is configured to use first and second weights with equal values when the sub-pixels having co-ordinates with even values of K and L are equidistant from the sub-pixel having co-ordinates with odd values of K and L whose value is being interpolated.

38. An interpolator according to claim 33, wherein the interpolator is configured to interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a weighted sum of the values of sub-pixels having co-ordinates with K equal to an even value and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions when the value of a sub-pixel having co-ordinates with K equal to an even value and L equal to an odd value is also required.

39. An interpolator according to claim 33, wherein the interpolator is configured to interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to an even value and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions when the value of a sub-pixel having co-ordinates with K equal to an odd value and L equal to an even value is also required.

40. A method according to claim 33, comprising truncating the second intermediate value by mathematically dividing the second intermediate value by a second scale factor, said second scale factor having a value equal to the sum of the respective weights used in the weighted sum of claim 63 multiplied by the sum of the respective weights used in the weighted sum of claim 67, thereby forming a sub-pixel value with a dynamic range equal to the dynamic range of said pixels.

41. An interpolator according to claim 33, wherein the interpolator is configured to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to 1 and L equal to zero, by taking an average of the value of a pixel having co-ordinates with K equal to zero and L equal to zero, and a sub-pixel having co-ordinates with K equal to 2 and L equal to zero.

42. An interpolator according to claim 33, wherein the interpolator is configured to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to 2^N−

1 and L equal to zero, by taking an average of the value of a pixel having co-ordinates with K equal to 2^N and L equal to zero, and a sub-pixel having co-ordinates with K equal to 2^N31 2 and L equal to zero.

43. An interpolator according to claim 33, wherein the interpolator is configured to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to 1 and L equal to 2^N, by taking an average of the value of a pixel having co-ordinates with K equal to zero and L equal to 2^N, and a sub-pixel having co-ordinates with K equal to 2 and L equal to 2^N.

44. An interpolator according to claim 33, wherein the interpolator is configured to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to 2^N−

1 and L equal to 2^N, by taking an average of the value of a pixel having co-ordinates with K equal to 2^N and L equal to 2^N, and a sub-pixel having co-ordinates with K equal to 2^N−

2 and L equal to 2^N.

45. An interpolator according to claim 33, wherein the interpolator is configured to interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an even value E, not including zero or 2^N, and L equal to an odd value O, by taking an average of the value of a first sub-pixel having co-ordinates with K equal to E and L equal to the even value O−

1 and a second sub-pixel having co-ordinates with K equal to E and L equal to the even value O+1.

46. An interpolator according to claim 33, wherein the interpolator is configured to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to zero and L equal to 1, by taking an average of the value of a pixel having co-ordinates with K equal to zero and L equal to zero, and a sub-pixel having co-ordinates with K equal to zero and L equal to 2.

47. An interpolator according to claim 33, wherein the interpolator is configured to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to 2^N and L equal to 1, by taking an average of the value of a pixel having co-ordinates with K equal to 2^N and L equal to zero, and a sub-pixel having co-ordinates with K equal to 2^N and L equal to 2.

48. An interpolator according to claim 33, wherein the interpolator is configured to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to zero and L equal to 2^N−

1, by taking an average of the value of a pixel having co-ordinates with K equal to zero and L equal to 2^N, and a sub-pixel having co-ordinates with K equal to zero and L equal to 2^N−

2.

49. An interpolator according to claim 33, wherein the interpolator is configured to interpolate a sub-pixel value for a sub-pixel having co-ordinates with K equal to 2^N and L equal to 2^N−

1, by taking an average of the value of a pixel having co-ordinates with K equal to 2^N and L equal to 2^N, and a sub-pixel having co-ordinates with K equal to 2^N and L equal to 2^N−

2.

50. An interpolator according to claim 33, wherein the interpolator is configured to interpolate a sub-pixel value for the sub-pixel having co-ordinates with K equal to 2^N−

1 and L equal to 2^N−

1, by taking a weighted average of the values of the four corner pixels that define the rectangular bounded region.

51. An interpolator according to claim 33, wherein N is set equal to 2.

52. An interpolator according to claim 33, wherein N is set equal to 3.

53. An interpolator according to claim 33, wherein the pixels have a predetermined dynamic range, said predetermined dynamic range corresponding to the range of values said pixels can take.

54. An interpolator according to claim 53, wherein the interpolator is configured to interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an even value and L equal zero and for sub-pixels having co-ordinates with K equal to zero and L equal to an even value, using weighted sums of the values of pixels located in rows and columns respectively, to produce a first intermediate value, said first intermediate value having a dynamic range equal to the predetermined dynamic range of the pixels used in said weighted sum multiplied by a value equal to the sum of the respective weights used in said weighted sum.

55. An interpolator according to claim 54, wherein the dynamic range of said first intermediate value is defined by the number of bits used to represent the respective pixels used in said weighted sum plus the number of bits required to represent the sum of the respective weights used in said weighted sum.

56. An interpolator according to claim 54, wherein the interpolator is configured to truncate the first intermediate value by mathematically dividing the first intermediate value by a first scale factor, said first scale factor having a value equal to the sum of the respective weights used in said weighted sum, thereby forming a sub-pixel value with a dynamic range equal to the predetermined dynamic range of said pixels.

57. An interpolator according to claim 56, wherein the interpolator is configured to use a truncated first intermediate sub-pixel value when interpolating a value for a sub-pixel having co-ordinates with odd values of K and L.

58. An interpolator according to claim 54, wherein the interpolator is configured to interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, using a weighted sum comprising first intermediate values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and corresponding first intermediate values for sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, to produce a second intermediate value, said second intermediate value having a dynamic range equal to the dynamic range of the first intermediate values used in said weighted sum, multiplied by a value equal to the sum of the respective weights used in said weighted sum.

59. An interpolator according to claim 58, wherein the interpolator is configured to truncate the second intermediate value by mathematically dividing the second intermediate value by a second scale factor, said second scale factor having a value equal to the sum of the respective weights used in the weighted sum of claim 54 multiplied by the sum of the respective weights used in the weighted sum of claim 58, thereby forming a sub-pixel value with a dynamic range equal to the predetermined dynamic range of said pixels.

60. An interpolator according to claim 59, wherein the interpolator is configured to use a truncated second intermediate sub-pixel value when interpolating a value for a sub-pixel having co-ordinates with odd values of K and L.

61. An interpolator according to claim 54, wherein the interpolator is configured to interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, using a weighted sum comprising first intermediate values for sub-pixels having co-ordinates with K equal to zero and L equal to an even value and corresponding first intermediate values for sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded region to produce a second intermediate value, said second intermediate value having a dynamic range equal to the dynamic range of the first intermediate values used in said weighted sum, multiplied by a value equal to the sum of the respective weights used in said weighted sum.

62. An interpolator according to claim 61, wherein the interpolator is configured to truncated the second intermediate value by mathematically dividing the second intermediate value by a second scale factor, said second scale factor having a value equal to the sum of the respective weights used in the weighted sum of claim 54 multiplied by the sum of the respective weights used in the weighted sum of claim 61, thereby forming a sub-pixel value with a dynamic range equal to the predetermined dynamic range of said pixels.

63. An interpolator according to claim 62, wherein the interpolator is configured to use a truncated second intermediate sub-pixel value when interpolating a value for a sub-pixel having co-ordinates with odd values of K and L.

64. A video encoder comprising an interpolator for sub-pixel value interpolation, the interpolator being configured to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically within the rectangular bounded region using the co-ordinate notation K/2^N, L/2^N, K and L being positive integers having respective values between zero and 2^N, N being a positive integer greater than one and representing a particular degree of sub-pixel value interpolation, the interpolator being configured to:

interpolate a sub-pixel value for a sub-pixel having co-ordinates with odd values of K and L, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates ½

, ½

, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with even values of K and L, including zero, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co-ordinates ½

, ½ and

the nearest neighbouring pixel;

interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to an even value, used in the interpolation of the sub-pixels having co-ordinates with odd values of K and L, using weighted sums of the values of pixels located in rows and columns respectively; and

interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a predetermined choice of either a weighted sum of the values of sub-pixels having co-ordinates with K equal to an even value and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, or a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to an even value and the values of sub-pixels having corresponding co-ordinates in immediately adjacent bounded rectangular regions.

65. A codec comprising a video encoder according to claim 64 and a video decoder comprising an interpolator for sub-pixel value interpolation, the interpolator being configured to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically within the rectangular bounded region using the co-ordinate notation K/2^N, L/2^N, K and L being positive integers having respective values between zero and 2^N, N being a positive integer greater than one and representing a particular degree of sub-pixel value interpolation, the interpolator being configured to:

interpolate a sub-pixel value for a sub-pixel having co-ordinates with odd values of K and L, according to a predetermined choice of a either weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates ½

, ½

, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with even values of K and L, including zero, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co- ordinates ½

, ½ and

the nearest neighbouring pixel;

interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to an even value, used in the interpolation of the sub-pixels having co-ordinates with odd values of K and L, using weighted sums of the values of pixels located in rows and columns respectively; and

interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a predetermined choice of either a weighted sum of the values of sub-pixels having co-ordinates with K equal to an even value and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, or a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to an even value and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions.

66. A communications terminal comprising a video encoder according to claim 64.

67. A communications terminal according to claim 66, wherein the communications terminal is a wireless terminal.

68. A still image encoder comprising an interpolator for sub-pixel value interpolation, the interpolator being configured to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically within the rectangular bounded region using the co-ordinate notation K/2^N, L/2^N, K and L being positive integers having respective values between zero and 2^N, N being a positive integer greater than one and representing a particular degree of sub-pixel value interpolation, the interpolator being configured to:

interpolate a sub-pixel value for a sub-pixel having co-ordinates with odd values of both K and L, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates ½

, ½

, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with even values of K and L, including zero, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co-ordinates ½

, ½ and

the nearest neighbouring pixel;

interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to an even value, used in the interpolation of the sub-pixels having co-ordinates with odd values of K and L, using weighted sums of the values of pixels located in rows and columns respectively; and

interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a predetermined choice of either a weighted sum of the values of sub-pixels having co-ordinates with K equal to an even value and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, or a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to an even value and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions.

69. A codec comprising a still image encoder according to claim 68 and a still image decoder comprising an interpolator for sub-pixel value interpolation, the interpolator being configured to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically within the rectangular bounded region using the co-ordinate notation K/2^N, L/2^N, K and L being positive integers having respective values between zero and 2^N, N being a positive integer greater than one and representing a particular degree of sub-pixel value interpolation, the interpolator being configured to:

interpolate a sub-pixel value for a sub-pixel having co-ordinates with odd values of K and L, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates ½

, ½

, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with even values of K and L, including zero, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co- ordinates ½

, ½ and

the nearest neighbouring pixel;

interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to an even value, used in the interpolation of the sub-pixels having co-ordinates with odd values of K and L, using weighted sums of the values of pixels located in rows and columns respectively; and

interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a predetermined choice of either a weighted sum of the values of sub-pixels having co-ordinates with K equal to an even value and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, or a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to an even value and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions.

70. A communications terminal comprising a still image encoder according to claim 68.

71. A video encoder comprising an interpolator for sub-pixel value interpolation, the interpolator being configured to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically within the rectangular bounded region using the co-ordinate notation K/2^N, L/2^N, K and L being positive integers having respective values between zero and 2^N, N being a positive integer greater than one and representing a particular degree of sub-pixel value interpolation, the interpolator being configured to:

interpolate a sub-pixel value for a sub-pixel having co-ordinates with odd values of K and L, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates ½

, ½

, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with even values of K and L, including zero, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co-ordinates ½

, ½ and

the nearest neighbouring pixel;

interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to an even value, used in the interpolation of the sub-pixels having co-ordinates with odd values of K and L, using weighted sums of the values of pixels located in rows and columns respectively; and

interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a predetermined choice of either a weighted sum of the values of sub-pixels having co-ordinates with K equal to an even value and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, or a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to an even value and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded region.

72. A communications terminal comprising a video decoder according to claim 71.

73. A still image decoder comprising an interpolator for sub-pixel value interpolation, the interpolator being configured to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically within the rectangular bounded region using the co-ordinate notation K/2^N, L/2^N, K and L being positive integers having respective values between zero and 2^N, N being a positive integer greater than one and representing a particular degree of sub-pixel value interpolation, the interpolator being configured to:

interpolate a sub-pixel value for a sub-pixel having co-ordinates with odd values of K and L, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates ½

, ½

, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with even values of K and L, including zero, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co-ordinates ½

, ½ and

the nearest neighbouring pixel;

interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to an even value, used in the interpolation of the sub-pixels having co-ordinates with odd values of K and L, using weighted sums of the values of pixels located in rows and columns respectively; and

interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a predetermined choice of either a weighted sum of the values of sub-pixels having co-ordinates with K equal to an even value and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, or a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to an even value and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions.

74. A communications terminal comprising a still image decoder according to claim 73.

75. An interpolator for sub-pixel value interpolation, the interpolator being configured to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically within the rectangular bounded region using the co-ordinate notation K/2^N, L/2^N, K and L being positive integers having respective values between zero and 2^N, N being a positive integer greater than one and representing a particular degree of sub-pixel value interpolation, the interpolator comprising:

circuitry configured to interpolate a sub-pixel value for a sub-pixel having co-ordinates with odd values of K and L, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates ½

, ½

, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with even values of K and L, including zero, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co-ordinates ½

, ½ and

the nearest neighbouring pixel;

circuitry configured to interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to an even value, used in the interpolation of the sub-pixels having co-ordinates with odd values of K and L, using weighted sums of the values of pixels located in rows and columns respectively; and

circuitry configured to interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a predetermined choice of either a weighted sum of the values of sub-pixels having co-ordinates with K equal to an even value and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, or a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to an even value and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions.

76. An interpolator for quarter resolution sub-pixel value interpolation, the interpolator being configured to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically within the rectangular bounded region using the co-ordinate notation K/4, L/4, K and L being positive integers having respective values between zero and 4, the interpolator being configured to:

interpolate a sub-pixel value for a sub-pixel having co-ordinates with K and L equal to 1 or 3, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates 2/4, 2/4, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with K and L equal to zero, 2 or 4, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co-ordinates 2/4, 2/4 and the nearest neighbouring pixel;

interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to 2 and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to 2, used in the interpolation of the sub-pixels having co-ordinates with K and L equal to 1 or 3, using weighted sums of the values of pixels located in rows and columns respectively; and

interpolate a sub-pixel value for the sub-pixel having co-ordinates with both K and L equal to 2, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with both K and L equal to 1 or 3, using a predetermined choice of either a weighted sum of the values of the sub-pixel having co-ordinates with K equal to 2 and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, or a weighted sum of the value of the sub-pixel having co-ordinates with K equal to zero and L equal to 2 and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions.

77. A video encoder or still image encoder comprising an interpolator according to claim 76.

78. A communications terminal comprising a video encoder or still image encoder according to claim 77.

79. An interpolator for eighth resolution sub-pixel value interpolation, the interpolator being configured to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically with the rectangular bounded region using the co-ordinate notation K/8, L/8, K and L being positive integers having respective values between zero and 8, the interpolator being configured to:

interpolate a sub-pixel value for a sub-pixel having co-ordinates with K and L equal to 1, 3, 5 or 7, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates 4/8, 4/8, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with both K and L equal to zero, 2, 4, 6 or 8, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co-ordinates 4/8, 4/8 and the nearest neighbouring pixel;

interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to 2, 4 or 6 and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to 2, 4 or 6, used in the interpolation of the sub-pixels having co-ordinates with both K and L equal to 1, 3, 5 or 7, using weighted sums of the values of pixels located in rows and columns respectively; and

interpolate sub-pixel values for sub-pixels having co-ordinates with K and L equal to 2, 4 or 6, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with K and L equal to 1, 3, 5 or 7, using a predetermined choice of either a weighted sum of the values of sub-pixels having co-ordinates with K equal to 2, 4 or 6 and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions or a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to 2, 4 or 6 and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions.

80. A video decoder or still image decoder comprising an interpolator according to claim 79.

81. A communications terminal comprising a video decoder or still image decoder according to claim 80.

82. A computer program for sub-pixel value interpolation, embodied on a computer readable storage medium, the computer program being configured to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically within the rectangular bounded region using the co-ordinate notation K/2^N, L/2^N, K and L being positive integers having respective values between zero and 2^N, N being a positive integer greater than one and representing a particular degree of sub-pixel value interpolation, the computer program comprising:

computer program code for interpolating a sub-pixel value for a sub-pixel having co-ordinates with odd values of both K and L, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates ½

, ½

, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with even values of both K and L, including zero, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co-ordinates ½

, ½ and

the nearest neighbouring pixel;

computer program code for interpolating sub-pixel values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to an even value, used in the interpolation of the sub-pixels having co-ordinates with odd values of K and L, using weighted sums of the values of pixels located in rows and columns respectively; and

computer program code for interpolating sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a predetermined choice of either a weighted sum of the values of sub-pixels having co-ordinates with K equal to an even value and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, or a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to an even value and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions.

83. A network apparatus comprising an interpolator for sub-pixel value interpolation, the interpolator being configured to determine values for sub-pixels situated within a rectangular bounded region defined by four corner pixels with no intermediate pixels between the corners, the pixels and sub-pixels being configured for display in rows and columns, pixel and sub-pixel locations in the rows and columns being representable mathematically within the rectangular bounded region using the co-ordinate notation K/2^N, L/2^N, K and L being positive integers having respective values between zero and 2^N, N being a positive integer greater than one and representing a particular degree of sub-pixel value interpolation, the interpolator being configured to:

interpolate a sub-pixel value for a sub-pixel having co-ordinates with odd values of K and L, according to a predetermined choice of either a weighted average of the value of a nearest-neighbouring pixel and the value of the sub-pixel situated at co-ordinates ½

, ½

, or a weighted average of the values of a pair of diagonally-opposed sub-pixels having co-ordinates with even values of K and L, including zero, situated within a quadrant of the rectangular bounded region, the quadrant being defined by the sub-pixel having co-ordinates ½

, ½ and

the nearest neighbouring pixel;

interpolate sub-pixel values for sub-pixels having co-ordinates with K equal to an even value and L equal to zero and sub-pixels having co-ordinates with K equal to zero and L equal to an even value, used in the interpolation of the sub-pixels having co-ordinates with odd values of K and L, using weighted sums of the values of pixels located in rows and columns respectively; and

interpolate sub-pixel values for sub-pixels having co-ordinates with even values of K and L, used in the interpolation of sub-pixel values for the sub-pixels having co-ordinates with odd values of K and L, using a predetermined choice of either a weighted sum of the values of sub-pixels having co-ordinates with K equal to an even value and L equal to zero and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions, or a weighted sum of the values of sub-pixels having co-ordinates with K equal to zero and L equal to an even value and the values of sub-pixels having corresponding co-ordinates in immediately adjacent rectangular bounded regions.