Adaptive coefficient scanning for video coding

• US 8,488,668 B2
• Filed: 06/04/2008
• Issued: 07/16/2013
• Est. Priority Date: 06/15/2007
• Status: Active Grant

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Ye, Yan, Karczewicz, Marta
• Primary Examiner(s): Sefcheck, Gregory
• Assistant Examiner(s): YOUNG, STEVE R
• Application Number:
  US12/133,277
• Publication Number:
  US 20080310504A1
• Time in Patent Office: 1,868 Days
• Field of Search: None
• US Class Current: 375/240.02
• CPC Class Codes: H04N 19/103 Selection of coding mode or of prediction mode
• CPC Class Codes: H04N 19/11 among a plurality of spatial predictive coding modes
• CPC Class Codes: H04N 19/12 Selection from among a plurality of transforms or standards, e.g. selection between discrete cosine transform [DCT] and sub-band transform or selection between H.263 and H.264
• 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 19/129 Scanning of coding units, e.g. zig-zag scan of transform coefficients or flexible macroblock ordering [FMO]
• CPC Class Codes: H04N 19/13 Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC]
• CPC Class Codes: H04N 19/147 according to rate distortion criteria rate-distortion as a criterion for motion estimation H04N19/567
• CPC Class Codes: H04N 19/157 Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/18 the unit being a set of transform coefficients
• CPC Class Codes: H04N 19/19 using optimisation based on Lagrange multipliers
• CPC Class Codes: H04N 19/196 being specially adapted for the computation of encoding parameters, e.g. by averaging previously computed encoding parameters processing of motion vectors H04N19/513
• CPC Class Codes: H04N 19/197 including determination of the initial value of an encoding parameter H04N19/56 takes precedence
• CPC Class Codes: H04N 19/42 characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation H04N19/635 takes precedence
• CPC Class Codes: H04N 19/46 Embedding additional information in the video signal during the compression process H04N19/517, H04N19/68, H04N19/70 take precedence
• CPC Class Codes: H04N 19/463 by compressing encoding parameters before transmission
• CPC Class Codes: H04N 19/48 using compressed domain processing techniques other than decoding, e.g. modification of transform coefficients, variable length coding [VLC] data or run-length data motion estimation in a transform domain H04N19/547; processing of decoded motion vectors H04N19/513
• CPC Class Codes: H04N 19/593 involving spatial prediction techniques
• CPC Class Codes: H04N 19/61 in combination with predictive coding
• CPC Class Codes: H04N 19/625 using discrete cosine transform [DCT]
• CPC Class Codes: H04N 19/70 : characterised by syntax aspects related to video coding, e.g. related to compression standards

Abstract

This disclosure describes techniques for scanning coefficients of video blocks. In particular, the techniques of this disclosure adapt a scan order used to scan a two-dimensional block of coefficients into a one-dimensional coefficient vector based on statistics associated with one or more previously coded blocks. For example, statistics that indicate the likelihood that a given coefficient value in each position of a two-dimensional block is zero or non-zero may be collected for one or more previously coded blocks. At some point, an adjustment to the scan order can be made in order to better ensure that non-zero coefficients are grouped together near the front of the one-dimensional coefficient vector, which can improve the effectiveness of entropy coding. The collection of statistics and adjustment of scan order may be made separately for each possible prediction mode.

Claims

1. A method of coding video data, the method comprising:

for each of first and second video blocks of a plurality of video blocks, generating a prediction block for the respective video block based on a first prediction mode of a plurality of prediction modes and one or more neighboring video blocks, each of the plurality of prediction modes being associated with a respective coefficient scan order of a plurality of coefficient scan orders;

for each of third and fourth video blocks of the plurality of video blocks, generating a prediction block for the respective video block based on a second prediction mode of the plurality of prediction modes and one or more neighboring video blocks, the second prediction mode being different than the first prediction mode;

for each of the plurality of video blocks, generating a reconstructed video block for the respective video block based on the prediction block for the respective video block and a residual block for the respective video block;

for each of the plurality of video blocks, applying one or more transforms to the respective video block to transform between residual pixel values of the respective video block and residual transform coefficient values of the respective video block, for each of the plurality of video blocks, selecting a coefficient scan order for scanning the respective video block from the plurality of coefficient scan orders based on a prediction mode used to generate the prediction block for the respective video block, the prediction mode being one of the plurality of prediction modes;

for each of the first and third video blocks, scanning the residual transform coefficient values of the respective video block using the coefficient scan order selected for the respective video block;

for each of the first and second prediction modes, collecting coefficient statistics for one or more of the video blocks within the plurality of video blocks that are associated with the respective prediction mode;

for each of the first and second prediction modes, adjusting the coefficient scan order for the respective prediction mode in accordance with the collected coefficient statistics for the respective prediction mode to generate an adjusted coefficient scan order for the respective prediction mode;

for each of the second and fourth video blocks, scanning the residual transform coefficient values of the respective video block using the adjusted coefficient scan order that corresponds to the prediction mode of the respective video block; and

for each of the plurality of video blocks, entropy coding the residual transform coefficient values for the respective video block.

2. The method of claim 1, wherein adjusting the coefficient scan order comprises, for at least one of the first and second prediction modes, adjusting the coefficient scan order for the respective prediction mode after a fixed interval to generate the adjusted coefficient scan order for the respective prediction mode.

3. The method of claim 2, wherein adjusting the coefficient scan order for the respective prediction mode after a fixed interval comprises, for the at least one of the first and second prediction modes, adjusting the coefficient scan order for the respective prediction mode after a threshold number of video blocks have been coded.

4. The method of claim 1, wherein collecting the coefficient statistics comprises, for each of the first and second prediction modes, maintaining a count value corresponding to each coefficient position within a portion of a two-dimensional representation of residual transform coefficient values for the respective prediction mode that tracks a number of times that each of the coefficient positions within the portion of the two-dimensional representation of the residual transform coefficient values has a non-zero coefficient value for the respective prediction mode.

5. The method of claim 4, wherein adjusting the coefficient scan order comprises, for each of the first and second prediction modes, adjusting the coefficient scan order for the respective prediction mode to generate an adjusted coefficient scan order for the respective prediction mode that scans the coefficient positions in descending order based on count values of the coefficient positions for the respective prediction mode.

6. The method of claim 4, wherein adjusting the coefficient scan order comprises, for each of the first and second prediction modes, updating the coefficient scan order for the respective prediction mode in accordance with the collected coefficient statistics for the respective prediction mode to generate the adjusted coefficient scan order for the respective prediction mode when one of the count values for the respective prediction mode reaches a threshold value.

7. The method of claim 1, further comprising normalizing collected coefficient statistics associated with at least one of the first and second prediction modes.

8. The method of claim 7, wherein:

collecting the coefficient statistics comprises, for the at least one of the first and second prediction modes, maintaining a count value corresponding to each coefficient position within a portion of a two-dimensional representation of residual transform coefficient values that tracks a number of times that each of the coefficient positions within the portion of the two-dimensional representation of the residual transform coefficient values has a non-zero coefficient value for the respective prediction mode; and

normalizing the collected coefficient statistics comprises, for the at least one of the first and second prediction modes, normalizing the collected coefficient statistics for the respective prediction mode when one of the count values for the respective prediction mode reaches a threshold count value.

9. The method of claim 7, wherein normalizing the collected coefficient statistics comprises one of reducing each of the count values for the respective prediction mode by a factor and resetting the count values for the respective prediction mode to a set of initial count values for the respective prediction mode.

10. The method of claim 1,

wherein collecting the coefficient statistics comprises collecting first coefficient statistics associated with the first prediction mode and collecting second coefficient statistics associated with the second prediction mode, and

wherein adjusting the coefficient scan order comprises adjusting a first coefficient scan order associated with the first prediction mode in accordance with the first coefficient statistics, and adjusting a second coefficient scan order associated with the second prediction mode in accordance with the second coefficient statistics.

11. The method of claim 1, wherein:

coding the video data comprises encoding the video data;

applying the one or more transforms comprises, for each of the plurality of video blocks, applying the one or more transforms to the respective video block to transform the residual pixel values of the respective video block to the residual transform coefficient values of the respective video block;

scanning the residual transform coefficient values comprises, for each of the plurality of video blocks, generating a one-dimensional vector of residual transform coefficient values for the respective video block from a two-dimensional block of residual transform coefficient values for the respective video block based on a coefficient scan order for the respective video block; and

entropy coding comprises, for each of the plurality of video blocks, entropy encoding the one-dimensional vector of residual transform coefficient values for the respective video block.

12. The method of claim 1, wherein:

coding the video data comprises decoding the video data;

entropy coding comprises, for each of the plurality of video blocks, entropy decoding a one-dimensional vector of residual transform coefficient values for the respective video block;

scanning the residual transform coefficient values comprises, for each of the plurality of video blocks, generating a two-dimensional block of residual transform coefficient values for the respective video block from the one-dimensional vector of residual transform coefficient values for the respective video block based on a coefficient scan order for the respective video block; and

applying the one or more transforms comprises, for each of the plurality of video blocks, applying the one or more transforms to the respective video block to transform the residual transform coefficient values of the respective video block to the residual pixel values of the respective video block.

13. The method of claim 1, wherein the first video block and the second video block belong to a first coding unit, the method further comprising re-initializing a coefficient scan order for scanning a second coding unit.

14. A device that codes video data, the device comprising:

a processor configured to;

for each of first and second video blocks of a plurality of video blocks, generate a prediction block for the respective video block based on a first prediction mode of a plurality of prediction modes and one or more neighboring video blocks, each of the plurality of prediction modes being associated with a respective coefficient scan order of a plurality of coefficient scan orders;

for each of third and fourth video blocks of the plurality of video blocks, generate a prediction block for the respective video block based on a second prediction mode of the plurality of prediction modes and one or more neighboring video blocks, the second prediction mode being different than the first prediction mode;

for each of the plurality of video blocks, generate a reconstructed video block for the respective video block based on the prediction block for the respective video block and a residual block for the respective video block;

for each of the plurality of video blocks, apply one or more transforms to the respective video block to transform between residual pixel values of the respective video block and residual transform coefficient values of the respective video block, for each of the plurality of video blocks, select a coefficient scan order for scanning the respective video block from the plurality of coefficient scan orders based on a prediction mode used to generate the prediction block for the respective video block, the prediction mode being one of the plurality of prediction modes;

for each of the first and third video blocks, scan the residual transform coefficient values of the respective video block using the coefficient scan order selected for the respective video block;

for each of the first and second prediction modes, collect coefficient statistics for one or more of the video blocks within the plurality of video blocks that are associated with the respective prediction mode;

for each of the first and second prediction modes, adjust the coefficient scan order for the respective prediction mode in accordance with the collected coefficient statistics for the respective prediction mode to generate an adjusted coefficient scan order for the respective prediction mode;

for each of the second and fourth video blocks, scan the residual transform coefficient values of the respective video block using the adjusted coefficient scan order that corresponds to the prediction mode of the respective video block; and

for each of the plurality of video blocks, entropy code the residual transform coefficient values for the respective video block.

15. The device of claim 14, wherein the processor is further configured to, for at least one of the first and second prediction modes, adjust the coefficient scan order for the respective prediction mode after a fixed interval to generate the adjusted coefficient scan order for the respective prediction mode.

16. The device of claim 15, wherein the processor is further configured to, for the at least one of the first and second prediction modes, adjust the coefficient scan order for the respective prediction mode after a threshold number of video blocks have been coded.

17. The device of claim 14, wherein the processor is further configured to, for each of the first and second prediction modes, maintain a count value corresponding to each coefficient position within a portion of a two-dimensional representation of residual transform coefficient values for the respective prediction mode that tracks a number of times that each of the coefficient positions within the portion of the two-dimensional representation of the residual transform coefficient values has a non-zero coefficient value for the respective prediction mode.

18. The device of claim 17, wherein the processor is further configured to, for each of the first and second prediction modes, adjust the coefficient scan order for the respective prediction mode to generate an adjusted coefficient scan order for the respective prediction mode that scans the coefficient positions in descending order based on count values of the coefficient positions for the respective prediction mode.

19. The device of claim 17, wherein the processor is further configured to, for each of the first and second prediction modes, update the coefficient scan order for the respective prediction mode in accordance with the collected coefficient statistics for the respective prediction mode to generate the adjusted coefficient scan order for the respective prediction mode when one of the count values for the respective prediction mode reaches a threshold value.

20. The device of claim 14, wherein the processor is further configured to normalize collected coefficient statistics associated with at least one of the first and second prediction modes.

21. The device of claim 20, wherein the processor is further configured to, for the at least one of the first and second prediction modes, maintain a count value corresponding to each coefficient position within a portion of a two-dimensional representation of residual transform coefficient values that tracks a number of times that each of the coefficient positions within the portion of the two-dimensional representation of the residual transform coefficient values has a non-zero coefficient value for the respective prediction mode and normalize the collected coefficient statistics for the respective prediction mode when one of the count values for the respective prediction mode reaches a threshold count value.

22. The device of claim 20, wherein the processor is further configured to one of reduce each of the count values for the respective prediction mode by a factor and reset the count values for the respective prediction mode to a set of initial count values for the respective prediction mode.

23. The device of claim 14, wherein the processor is further configured to:

collect first statistics associated with the first prediction mode;

collect second coefficient statistics associated with the second prediction mode;

adjust a first coefficient scan order associated with the first prediction mode in accordance with the first coefficient statistics; and

adjust a second coefficient scan order associated with the second prediction mode in accordance with the second coefficient statistics.

24. The device of claim 14,

wherein the device comprises an encoder device, and

wherein the processor is further configured to:

for each of the plurality of video blocks, apply the one or more transforms to the respective video block to transform the residual pixel values of the respective video block to the residual transform coefficient values of the respective video block;

for each of the plurality of video blocks, generate a one-dimensional vector of residual transform coefficient values for the respective video block from a two-dimensional block of residual transform coefficient values for the respective video block based on a coefficient scan order for the respective video block; and

for each of the plurality of video blocks, entropy encode the one-dimensional vector of residual transform coefficient values for the respective video block.

25. The device of claim 14,

wherein the device comprises a decoder device, and

wherein the processor is further configured to:

for each of the plurality of video blocks, entropy decode a one-dimensional vector of residual transform coefficient values for the respective video block;

for each of the plurality of video blocks, generate a two-dimensional block of residual transform coefficient values for the respective video block from a one-dimensional vector of residual transform coefficient values for the respective video block based on a coefficient scan order for the respective video block; and

for each of the plurality of video blocks, apply the one or more transforms to the respective video block to transform the residual transform coefficient values of the respective video block to the residual pixel values of the respective video block.

26. The device of claim 14, wherein the first video block and the second video block belong to a first coding unit and the processor is further configured to re-initialize a coefficient scan order for scanning a second coding unit.

27. The device of claim 14, wherein the device comprises a wireless communication device having a display, the display being configured to display the video data.

28. The device of claim 14, wherein the device comprises an integrated circuit device.

29. A non-transitory computer-readable medium comprising instructions that upon execution in a video coding device cause the device to code video blocks, wherein the instructions cause the device to:

for each of first and second video blocks of a plurality of video blocks, generate a prediction block for the respective video block based on a first prediction mode of a plurality of prediction modes and one or more neighboring video blocks, each of the plurality of prediction modes being associated with a respective coefficient scan order of a plurality of coefficient scan orders;

for each of third and fourth video blocks of the plurality of video blocks, generate a prediction block for the respective video block based on a second prediction mode of the plurality of prediction modes and one or more neighboring video blocks, the second prediction mode being different than the first prediction mode;

for each of the plurality of video blocks, generate a reconstructed video block for the respective video block based on the prediction block for the respective video block and a residual block for the respective video block;

for each of the plurality of video blocks, apply one or more transforms to the respective video block to transform between residual pixel values of the respective video block and residual transform coefficient values of the respective video block, for each of the plurality of video blocks, select a coefficient scan order for scanning the respective video block from the plurality of coefficient scan orders based on a prediction mode used to generate the prediction block for predicting the respective video block, the prediction mode being one of the plurality of prediction modes;

for each of the first and third video blocks, scan the residual transform coefficient values of the respective video block using the coefficient scan order selected for the respective video block;

for each of the first and second prediction modes, collect coefficient statistics for one or more of the video blocks within the plurality of video blocks that are associated with the respective prediction mode;

for each of the first and second prediction modes, adjust the coefficient scan order for the respective prediction mode in accordance with the collected coefficient statistics for the respective prediction mode to generate an adjusted coefficient scan order for the respective prediction mode;

for each of the second and fourth video blocks, scan the residual transform coefficient values of the respective video block using the adjusted coefficient scan order that corresponds to the prediction mode of the respective video block; and

for each of the plurality of video blocks, entropy code the residual transform coefficient values for the respective video block.

30. The non-transitory computer-readable medium of claim 29, wherein the instructions cause the device to, for at least one of the first and second prediction modes, adjust the coefficient scan order for the respective prediction mode after a fixed interval to generate the adjusted coefficient scan order for the respective prediction mode.

31. The non-transitory computer-readable medium of claim 30, wherein the instructions cause the device to, for the at least one of the first and second prediction modes, adjust the coefficient scan order for the respective prediction mode after a threshold number of video blocks have been coded.

32. The non-transitory computer-readable medium of claim 29, wherein the instructions cause the device to, for each of the first and second prediction modes, maintain a count value corresponding to each coefficient position within a portion of a two-dimensional representation of residual transform coefficient values for the respective prediction mode that tracks a number of times that each of the coefficient positions within the portion of the two-dimensional representation of the residual transform coefficient values has a non-zero coefficient value for the respective prediction mode.

33. The non-transitory computer-readable medium of claim 32, wherein the instructions cause the device to, for each of the first and second prediction modes, adjust the coefficient scan order for the respective prediction mode to generate an adjusted coefficient scan order for the respective prediction mode that scans the coefficient positions in descending order based on count values of the coefficient positions for the respective prediction mode.

34. The non-transitory computer-readable medium of claim 32, wherein the instructions cause the device to, for each of the first and second prediction modes, update the coefficient scan order for the respective prediction mode in accordance with the collected coefficient statistics for the respective prediction mode to generate the adjusted coefficient scan order for the respective prediction mode when one of the count values for the respective prediction mode reaches a threshold value.

35. The non-transitory computer-readable medium of claim 29, wherein the instructions cause the device to normalize collected coefficient statistics associated with at least one of the first and second prediction modes.

36. The non-transitory computer-readable medium of claim 35, wherein the instructions cause the device to:

for the at least one of the first and second prediction modes, maintain a count value corresponding to each coefficient position within a portion of a two-dimensional representation of residual transform coefficient values that tracks a number of times that each of the coefficient positions within the portion of the two-dimensional representation of the residual transform coefficient values has a non-zero coefficient value for the respective prediction mode; and

for the at least one of the first and second prediction modes, normalize the collected coefficient statistics for the respective prediction mode when one of the count values for the respective prediction mode reaches a threshold count value.

37. The non-transitory computer-readable medium of claim 35, wherein the instructions cause the device to normalize the collected coefficient statistics by reducing each of the count values for the respective prediction mode by a factor or resetting the count values for the respective prediction mode to a set of initial count values for the respective prediction mode.

38. The non-transitory computer-readable medium of claim 29, wherein the instructions cause the device to:

collect first statistics associated with the first prediction mode;

collect second coefficient statistics associated with the second prediction mode;

adjust a first coefficient scan order associated with the first prediction mode in accordance with the first coefficient statistics; and

adjust a second coefficient scan order associated with the second prediction mode in accordance with the second coefficient statistics.

39. The non-transitory computer-readable medium of claim 29, wherein the device comprises an encoder device, and

wherein the instructions cause the encoder device to:

for each of the plurality of video blocks, apply the one or more transforms to the respective video block to transform the residual pixel values of the respective video block to the residual transform coefficient values of the respective video block;

for each of the plurality of video blocks, generate a one-dimensional vector of residual transform coefficient values for the respective video block from a two-dimensional block of residual transform coefficient values for the respective video block based on a coefficient scan order for the respective video block; and

for each of the plurality of video blocks, entropy encode the one-dimensional vector of residual transform coefficient values for the respective video block.

40. The non-transitory computer-readable medium of claim 29, wherein the device comprises a decoder device, and

wherein the instructions cause the decoder device to:

for each of the plurality of video blocks, entropy decode a one-dimensional vector of residual transform coefficient values for the respective video block;

for each of the plurality of video blocks, generate a two-dimensional block of residual transform coefficient values for the respective video block from a one-dimensional vector of residual transform coefficient values for the respective video block based on a coefficient scan order for the respective video block; and

for each of the plurality of video blocks, apply the one or more transforms to the respective video block to transform the residual transform coefficient values of the respective video block to the residual pixel values of the respective video block.

41. The non-transitory computer-readable medium of claim 29, wherein the first video block and the second video block belong to a first coding unit, and wherein the instructions cause the device to re-initialize a coefficient scan order for scanning a second coding unit.

42. A device that codes video data, the device comprising:

means for generating, for each of first and second video blocks of a plurality of video blocks, a prediction block for the respective video block based on a first prediction mode of a plurality of prediction modes and one or more neighboring video blocks, each of the plurality of prediction modes being associated with a respective coefficient scan order of a plurality of coefficient scan orders;

means for generating, for each of third and fourth video blocks of the plurality of video blocks, a prediction block for the respective video block based on a second prediction mode of the plurality of prediction modes and one or more neighboring video blocks, the second prediction mode being different than the first prediction mode;

means for generating, for each of the plurality of video blocks, a reconstructed video block for the respective video block based on the prediction block for the respective video block and a residual block for the respective video block;

means for applying, for each of the plurality of video blocks, one or more transforms to the respective video block to transform between residual pixel values of the respective video block and residual transform coefficient values of the respective video block;

means for selecting, for each of the plurality of video blocks, a coefficient scan order for scanning the respective video block from the plurality of coefficient scan orders based on a prediction mode used to generate the prediction block for the respective video block, the prediction mode being one of the plurality of prediction modes;

means for scanning, for each of the first and third video blocks, the residual transform coefficient values of the respective video block using the coefficient scan order selected for the respective video block;

means for collecting, for each of the first and second prediction modes, coefficient statistics for one or more of the video blocks within the plurality of video blocks that are associated with the respective prediction mode;

means for adjusting, for each of the first and second prediction modes, the coefficient scan order for the respective prediction mode in accordance with the collected coefficient statistics for the respective prediction mode to generate an adjusted coefficient scan order for the respective prediction mode;

wherein the residual transform coefficient scanning means scans, for each of the second and fourth video blocks, the residual transform coefficient values of the respective video block using the adjusted coefficient scan order that corresponds to the prediction mode of the respective video block; and

means for entropy coding, for each of the plurality of video blocks, the residual transform coefficient values for the respective video block.

43. The device of claim 42, wherein the adjusting means adjusts, for at least one of the first and second prediction modes, the coefficient scan order for the respective prediction mode after a fixed interval to generate the adjusted coefficient scan order for the respective prediction mode.

44. The device of claim 43, wherein the adjusting means adjusts, for the at least one of the first and second prediction modes, the coefficient scan order for the respective prediction mode after a threshold number of video blocks have been coded.

45. The device of claim 42, wherein the statistics collecting means maintains, for each of the first and second prediction modes, a count value corresponding to each coefficient position within a portion of a two-dimensional representation of residual transform coefficient values that tracks a number of times that each of the coefficient positions within the portion of the two-dimensional representation of the residual transform coefficient values has a non-zero coefficient value for the respective prediction mode.

46. The device of claim 45, wherein the adjusting means adjusts, for each of the first and second prediction modes, the coefficient scan order for the respective prediction mode to generate an adjusted coefficient scan order for the respective prediction mode that scans the coefficient positions in descending order based on count values of the coefficient positions for the respective prediction mode.

47. The device of claim 45, wherein the adjusting means updates, for each of the first and second prediction modes, the coefficient scan order for the respective prediction mode in accordance with the collected coefficient statistics for the respective prediction mode to generate the adjusted coefficient scan order for the respective prediction mode when one of the count values for the respective prediction mode reaches a threshold value.

48. The device of claim 42, further comprising means for normalizing collected coefficient statistics associated with at least one of the first and second prediction modes.

49. The device of claim 48, wherein

the statistics collecting means maintains, for the at least one of the first and second prediction modes, a count value corresponding to each coefficient position within a portion of a two-dimensional representation of residual transform coefficient values that tracks a number of times that each of the coefficient positions within the portion of the two-dimensional representation of the residual transform coefficient values has a non-zero coefficient value for the respective prediction mode;

and

the normalizing means normalizes, for the at least one of the first and second prediction modes, the collected coefficient statistics for the respective prediction mode when one of the count values reaches a threshold count value for the respective prediction mode.

50. The device of claim 48, wherein the normalizing means normalizes the collected coefficient statistics by one of reducing each of the count values for the respective prediction mode by a factor and resetting the count values for the respective prediction mode to a set of initial count values for the respective prediction mode.

51. The device of claim 42,

wherein the statistics collecting means collects first statistics associated with the first prediction mode and collects second coefficient statistics associated with the second prediction mode, and

wherein the adjusting means adjusts a first coefficient scan order associated with the first prediction mode in accordance with the first coefficient statistics and adjusts a second coefficient scan order associated with the second prediction mode in accordance with the second coefficient statistics.

52. The device of claim 42, wherein:

the device comprises an encoder device;

the applying means applies, for each of the plurality of video blocks, the one or more transforms to the respective video block to transform the residual pixel values of the respective video block to the residual transform coefficient values of the respective video block;

the scanning means generates, for each of the plurality of video blocks, a one-dimensional vector of residual transform coefficient values for the respective video block from a two-dimensional block of residual transform coefficient values for the respective video block based on a coefficient scan order for the respective video block; and

the entropy coding means entropy encodes, for each of the plurality of video blocks, the one-dimensional vector of residual transform coefficient values for the respective video block.

53. The device of claim 42, wherein:

the device comprises a decoder device;

the entropy coding means entropy decodes, for each of the plurality of video blocks, a one-dimensional vector of residual transform coefficient values for the respective video block;

the scanning means generates, for each of the plurality of video blocks, a two-dimensional block of residual transform coefficient values for the respective video block from a one-dimensional vector of residual transform coefficient values for the respective video block based on a coefficient scan order for the respective video block; and

the applying means applies, for each of the plurality of video blocks, the one or more transforms to the respective video block to transform the residual transform coefficient values of the respective video block to the residual pixel values of the respective video block.

54. The device of claim 42, wherein the first video block and the second video block belong to a first coding unit, and the scanning means re-initializes a coefficient scan order for scanning a second coding unit.