1. A method of encoding a unit of video data, the method comprising:dividing a transform block of video data into a plurality of transform coefficient sub-sets;determining, for each respective transform coefficient sub-set of the plurality of transform coefficient sub-sets, whether or not to generate a respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients; and generating the respective syntax element for each of the transform coefficient sub-sets of the plurality of transform coefficient sub-sets for which it was determined to generate the syntax element.

2. The method of claim 1, wherein determining, for the respective syntax element sub-set, whether or not to generate the respective syntax element comprises determining whether generating the respective syntax element will improve coding efficiency.

3. The method of claim 1, wherein determining whether or not to generate the respective syntax element comprises comparing a number of potential non-zero coefficients of the respective syntax element sub-set to at least one threshold; and if the number of potential non-zero coefficients of the respective syntax element sub-set is greater than the threshold, generating the respective syntax element.

4. The method of claim 3, further comprising:determining the number of potential non-zero coefficients based on how many coefficients of the respective syntax element sub-set have a position after a last non-zero coefficient of the transform block in a scan order.

5. The method of claim 1, wherein determining whether or not to generate the respective syntax element comprises:determining, for a plurality of previously coded transform blocks of video data, an average number of non-zero coefficients included in the respective syntax element sub-set;comparing the average number of non-zero coefficients included in the respective syntax element sub-set to a threshold; and if the average number of non-zero coefficients is less than the threshold, generating the respective syntax element.

6. The method of claim 1, wherein determining whether or not to generate the respective syntax element comprises:determining whether the respective syntax element sub-set includes a last non-zero coefficient for the transform block of video data in a scan order, wherein the respective syntax element that indicates whether or not the respective syntax element sub-set includes any non-zero coefficients is not generated if the respective syntax element sub-set includes the last non-zero coefficient.

7. A device configured to encode at least one transform block of video data, the device comprising:a memory configured to store video data; and one or more processors configured to:divide a transform block of video data into a plurality of transform coefficient sub-sets;determine, for each respective transform coefficient sub-set of the plurality of transform coefficient sub-sets, whether or not to generate a respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients; and generate the respective syntax element for each of the transform coefficient sub-sets of the plurality of transform coefficient sub-sets for which it was determined to generate the syntax element.

8. The device of claim 7, wherein the one or more processors are further configured to:determine, for the respective transform coefficient sub-set, whether or not to generate the respective syntax element based on whether generating the respective syntax element will improve coding efficiency.

9. The device of claim 7, wherein the one or more processors are further configured to:determine whether or not to generate the respective syntax element based on comparing a number of potential non-zero coefficients of the respective transform coefficient sub-set to at least one threshold; and if the number of potential non-zero coefficients of the respective transform coefficient sub-set is greater than the threshold, generate the respective syntax element

10. The device of claim 9, wherein the one or more processors are further configured to:determine the number of potential non-zero coefficients based on how many coefficients of the respective transform coefficient sub-set have a position after a last non-zero coefficient of the transform block in a scan order

11. The device of claim 7, wherein the one or more processors are further configured to:determine, for a plurality of previously coded transform blocks of video data, an average number of non-zero coefficients included in the respective transform coefficient sub-set; compare the average number of non-zero coefficients included in the respective transform coefficient sub-set to a threshold; and if the average number of non-zero coefficients is less than the threshold, generate the respective syntax element

12. The device of claim 7, wherein the one or more processors are further configured to:determining whether the respective transform coefficient sub-set includes a last non-zero coefficient for the transform block of video data in a scan order, wherein the respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients is not generated if the respective transform coefficient sub-set includes the last non-zero coefficient

13. A computer-readable storage medium that stores instructions that, when executed, cause a computing device to:divide a transform block of video data into a plurality of transform coefficient sub-sets;determine, for each respective transform coefficient sub-set of the plurality of transform coefficient sub-sets, whether or not to generate a respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients; and generate the respective syntax element for each of the transform coefficient sub-sets of the plurality of transform coefficient sub-sets for which it was determined to generate the syntax element

14. The computer-readable storage medium of claim 13, wherein the instructions further cause the computing device to: determine, for the respective transform coefficient sub-set, whether or not to generate the respective syntax element based on whether generating the respective syntax element will improve coding efficiency

15. The computer-readable storage medium of claim 13, wherein the instructions further cause the computing device to:determine whether or not to generate the respective syntax element based on comparing a number of potential non-zero coefficients of the respective transform coefficient sub-set to at least one threshold; and if the number of potential non-zero coefficients of the respective transform coefficient sub-set is greater than the threshold, generate the respective syntax element

16. The computer-readable storage medium of claim 15, wherein the instructions further cause the computing device to:determine the number of potential non-zero coefficients based on how many coefficients of the respective transform coefficient sub-set have a position after a last non-zero coefficient of the transform block in a scan order

17. The computer-readable storage medium of claim 13, wherein the instructions further cause the computing device to:determine, for a plurality of previously coded transform blocks of video data, an average number of non-zero coefficients included in the respective transform coefficient sub-set;compare the average number of non-zero coefficients included in the respective transform coefficient sub-set to a threshold; and if the average number of non-zero coefficients is less than the threshold, generate the respective syntax element

18. The computer-readable storage medium of claim 13, wherein the instructions further cause the computing device to:determine whether the respective transform coefficient sub-set includes a last non-zero coefficient for the transform block of video data in a scan order, wherein the respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients is not generated if the respective transform coefficient sub-set includes the last non-zero coefficient.19. A device configured to encode at least one transform block of video data, comprising:means for dividing a transform block of video data into a plurality of transform coefficient sub-sets;means for determining, for each respective transform coefficient sub-set of the plurality of transform coefficient sub-sets, whether or not to generate a syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients; and means for generating the respective syntax element for each of the transform coefficient sub-sets of the plurality of transform coefficient sub-sets for which it was determined to generate the syntax element.

20. The device of claim 19, further comprising:means for determining, for the respective transform coefficient sub-set, whether or not to generate the respective syntax element based on whether generating the respective syntax element will improve coding efficiency.

21. The device of claim 19, further comprising:means for determining whether or not to generate the respective syntax element based on comparing a number of potential non-zero coefficients of the respective transform coefficient sub-set to at least one threshold; and means for, if the number of potential non-zero coefficients of the respective transform coefficient sub-set is greater than the threshold, generating the respective syntax element.

22. The device of claim 21, further comprising:means for determining the number of potential non-zero coefficients based on how many coefficients of the respective transform coefficient sub-set have a position after a last non-zero coefficient of the transform block in a scan order.

23. The device of claim 19, wherein the means for determining whether or not to generate the respective syntax element comprise:means for determining, for a plurality of previously coded transform blocks of video data, an average number of non-zero coefficients included in the respective transform coefficient sub-set;means for comparing the average number of non-zero coefficients included in the respective transform coefficient sub-set to a threshold; and means for, if the average number of non-zero coefficients is less than the threshold, generating the respective syntax element.

24. The device of claim 19, further comprising:means for determining whether the respective transform coefficient sub-set includes a last non-zero coefficient for the transform block of video data in a scan order, wherein the respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients is not generated if the respective transform coefficient sub-set includes the last non-zero coefficient.

25. A method of decoding a unit of video data, comprising:receiving an encoded video bitstream including syntax elements for a transform block of video data;dividing the transform block of video data into a plurality of transform coefficient sub-sets;determining, for each respective transform coefficient sub-set of the plurality of transform coefficient sub-sets, whether the syntax elements included in the encoded video bitstream include a respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients; and determining, for each respective transform coefficient sub-set, whether or not to decode the respective transform coefficient sub-set based on the respective element.

26. The method of claim 25, further comprising:if the syntax elements included in the encoded video bitstream do not include the respective syntax element for the respective transform coefficient sub-set, decoding the respective transform coefficient sub-set.

27. The method of claim 26, further comprising:if the syntax elements included in the encoded video bitstream do include the respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficents, determining whether to decode the respective transform coefficient sub-set based on the indication of the respective syntax element.

28. The method of claim 27, further comprising: if the respective syntax element for the respective transform coefficient sub-set has a first value, decoding the respective transform coefficient sub-set; and if the respective syntax element for the respective transform coefficient sub-set has a second value different than the first value, not decoding the respective transform coefficient sub-set.

29. The method of claim 25, further comprising:if the syntax elements included in the encoded video bitstream do not include the respective syntax element for the respective transform coefficient sub-set, determining whether the encoded video bitstream includes a separate syntax element that indicates that a last non-zero coefficient in a scan order is included in the respective transform coefficient sub-set; and decoding the respective transform coefficient sub-set if the coded video bitstream includes the separate syntax element that indicates that a last non-zero coefficient in the scan order is included in the respective transform coefficient sub-set.

30. A device configured to decode a unit of video data, the device comprising: a memory configured to store a transform block of video data; and one or more processors configured to:receive an encoded video bitstream including syntax elements for the transform block of video data;divide the transform block of video data into a plurality of transform coefficient sub-sets;determine, for each respective transform coefficient sub-set of the plurality of transform coefficient sub-sets, whether the syntax elements included in the encoded video bitstream include a respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients; and determine, for each respective transform coefficient sub-set, whether or not to decode the respective transform coefficient sub-set based on the respective syntax element.

31. The device of claim 30, wherein the one or more processors are further configured to:if the syntax elements included in the encoded video bitstream do not include the respective syntax element for the respective transform coefficient sub-set, decode the respective transform coefficient sub-set.

32. The device of claim 31, wherein the one or more processors are further configured to:if the syntax elements included in the encoded video bitstream do include the respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients, determine whether to decode the respective transform coefficient sub-set based on the indication of the respective syntax element.

33. The device of claim 32, wherein the one or more processors are further configured to:if the respective syntax element for the respective transform coefficient sub-set has a first value, decode the respective transform coefficient sub-set; and if the respective syntax element for the respective transform coefficient sub-set has a second value different than the first value, not decode the respective transform coefficient sub-set.

34. The device of claim 30, wherein the one or more processors are further configured to:if the syntax elements included in the encoded video bitstream do not include the respective syntax element for the respective transform coefficient sub-set, determine whether the encoded video bitstream includes a separate syntax element that indicates that a last non-zero coefficient in a scan order is included in the respective transform coefficient sub-set; and decode the respective transform coefficient sub-set if the coded video bitstream includes the separate syntax element that indicates that a last non-zero coefficient in the scan order is included in the respective transform coefficient sub-set.

35. A computer-readable storage medium that stores instructions that, when executed, cause a computing device to:receive an encoded video bitstream including syntax elements for a transform block of video data;divide the transform block of video data into a plurality of transform coefficient sub-sets;determine, for each respective transform coefficient sub-set of the plurality of transform coefficient sub-sets, whether the syntax elements included in the encoded video bitstream include a respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients; and determine, for each respective transform coefficient sub-set, whether or not to decode the respective transform coefficient sub-set based on the respective syntax element.

36. The computer-readable storage medium of claim 35, wherein the instructions further cause the computing device to: if the syntax elements included in the encoded video bitstream do not include the respective syntax element for the respective transform coefficient sub-set, decode the respective transform coefficient sub-set.

37. The computer-readable storage medium of claim 36, wherein the instructions further cause the computing device to:if the syntax elements included in the encoded video bitstream do include the respective syntax element for the respective transform coefficient sub-set, determine whether to decode the respective transform coefficient sub-set based on the indication of the respective syntax element.

38. The computer-readable storage medium of claim 37, wherein the instructions further cause the computing device to:if the respective syntax element for the respective transform coefficient sub-set has a first value, decode the respective transform coefficient sub-set; and if the respective syntax element for the respective transform coefficient sub-set has a second value different than the first value, not decode the respective transform coefficient sub-set.

39. The computer-readable storage medium of claim 35, wherein the instructions further cause the computing device to:if the syntax elements included in the encoded video bitstream do not include the syntax element, determine whether the encoded video bitstream includes a separate syntax element that indicates that a last non-zero coefficient in a scan order is included in the respective transform coefficient sub-set; and decode the respective transform coefficient sub-set if the encoded video bitstream includes the separate syntax element that indicates that a last non-zero coefficient in the scan order is included in the respective transform coefficient sub-set.

40. A device configured to decode a unit of video data, comprising:means for receiving an encoded video bitstream including syntax elements encoding a transform block of video data;means for dividing the transform block of video data into a plurality of transform coefficient sub-sets;means for determining, for each respective transform coefficient sub-set of the plurality of transform coefficient sub-sets, whether the syntax elements included in the encoded video bitstream include a respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients; and means for determining, for each respective transform coefficient sub-set, whether or not to decode the respective transform coefficient sub-set based on the respective syntax element.

41. The device of claim 40, further comprising:means for, if the syntax elements included in the encoded video bitstream do not include the respective syntax element, decoding the respective transform coefficient sub-set.

42. The device of claim 41, further comprising:means for, if the syntax elements included in the encoded video bitstream do include the respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients, determining whether to decode the respective transform coefficient sub-set based on the indication of the respective syntax element.

43. The device of claim 42, further comprising:means for, if the respective syntax element for the respective transform coefficient sub-set has a first value, decoding the respective transform coefficient sub-set; and means for, if the respective syntax element for the respective transform coefficient sub-set has a second value different than the first value, not decoding the respective transform coefficient sub-set.

44. The device of claim 40, further comprising:if the syntax elements included in the encoded video bitstream do not include the respective syntax element for the respective transform coefficient sub-set, means for determining whether the encoded video bitstream includes a separate syntax element that indicates that a last non-zero coefficient in a scan order is included in the respective transform coefficient sub-set; and means for decoding the respective transform coefficient subset if the encoded video bitstream includes the separate syntax element that indicates that a last non-zero coefficient in the scan order is included in the respective transform coefficient sub-set.

45. The method of claim 1, wherein generating the respective syntax element for each of the transform coefficient sub-sets of the plurality of transform coefficient sub-sets for which it was determined to generate the syntax element comprises:generating the respective syntax element for fewer than all of the transform coefficient sub-sets of the plurality of transform coefficient sub-sets.

46. The method of claim 25, wherein determining, for each respective transform coefficient sub-set of the plurality of transform coefficient sub-sets, whether the syntax elements included in the encoded video bitstream include a respective syntax element that indicates whether or not the respective transform coefficient sub-set includes any non-zero coefficients comprises: determining that the syntax elements included in the encoded video bitstream include a respective syntax element for fewer than all of the transform coefficient sub-sets.']