Method and apparatus for rapid video and field coding

1. An apparatus for encoding video data for an image that is divisible into macroblocks and in which a plurality of macroblocks are assignable as a group of macroblocks, comprising:

a motion activity detector for receiving the image as a whole image; and

an encoder for assigning the group of macroblocks for encoding in one of a frame mode or a field mode responsive to application of the image to the motion activity detector,

wherein said group of macroblocks is a super-macroblock (super-MB), and

wherein the motion activity detector determines whether to encode in the frame mode or the field mode based on a percentage of super-MBs classified as moving regions in the image.

2. The apparatus according to claim 1 wherein the motion activity detector is a bandpass/high-pass filter.

3. The apparatus according to claim 2 , wherein a macroblock adaptive frame field (MBAFF) decision is responsive to application of the band-pass/high-pass filter to the image.

4. The apparatus according to claim 2 , wherein the super-MB comprises a plurality of pixels, and said encoder classifies at least some of the plurality of pixels as an edge pixel using an adaptive threshold for the band-pass/high-pass filter, determines a sum by adding a number of edge pixels, and compares the sum to another pre-specified threshold to determine whether the super-MB is classified with respect to the frame mode or the field mode.

5. An apparatus for encoding video data for an image that is divisible into super-macroblocks (super-MBs), comprising:

an encoder for performing a picture adaptive frame field (PAFF) decision using a band-pass/high-pass filter applied vertically to the super-MBs to respectively classify ones of the super-MBs as a moving region or a non-moving region in the image,

wherein the PAFF decision is based on a percentage of super-MBs classified as moving regions in the image.

6. The apparatus according to claim 5 , wherein the band-pass/high-pass filter uses an adaptable threshold that is selected to distinguish moving edges from true edges.

7. The apparatus according to claim 5 , wherein the PAFF decision is combined with other criteria comprising at least one selected from the group of frame type, absolute difference of a current picture and previous reference pictures, macroblock characteristics, frame variance, and macroblock variance.

8. The apparatus according to claim 5 , wherein the PAFF decision is performed on a Group Of Pictures (GOPs) level, for all frames in a current GOP, such that the field mode is used for the current GOP when a number of frames decided to be coded in the field mode is larger than a pre-specified percentage, otherwise the frame mode is used for the current GOP.

9. The apparatus according to claim 6 , wherein the threshold is adaptable frame-wise, Group Of Picture (GOP)-wise, or scene-wise.

10. The apparatus according to claim 6 , wherein the band-pass/high-pass filter is applied to one field image, and the adaptable threshold is selected where a number of edge pixels in the one field image is lower than another threshold.

11. The apparatus according to claim 9 , wherein said encoder is further used for attributing weights to the edge pixels according to at least one of position within a Super-MB and correlation with neighboring pixels prior to thresholding.

12. A video encoder for encoding video data for an image that is divisible into super-macroblocks (super-MBs), comprising an encoder for utilizing a pre-analysis decision of whether to classify an individual super-MB in the image with respect to a field mode or a frame mode, for both a picture adaptive frame field (PAFF) mode selection and a macroblock adaptive frame field (MBAFF) mode selection,

wherein the decision of whether to classify the individual super-MB in field mode or frame mode is based on exceeding a motion percentage threshold for the super-MB.

13. A method for encoding video data for an image that is divisible into macroblocks and in which a plurality of the macroblocks are assignable as a group of macroblocks, comprising:

detecting motion beyond a threshold for sections of the group of macroblocks; and

assigning the group of macroblocks for encoding in one of a frame mode or a field mode responsive to whether motion beyond the threshold is detected in the detecting step,

wherein said group of macroblocks is a super-macroblock (super-MB), and

wherein the assigning step determines whether to encode in the frame mode or the field mode based on a percentage of super-MBs classified as moving regions in the image.

14. A method for encoding video data for an image, comprising the step of classifying a super-macroblock (super-MB) in the image with respect to one of a frame mode or a field mode using a band-pass/high-pass filter applied vertically to the image,

wherein the step of classifying the super-MB in frame mode or field mode is based on exceeding a motion percentage threshold for the super-MB.

15. The method according to claim 14 , further comprising the step of utilizing a classification obtained by the band-pass/high-pass filter for a macroblock adaptive frame field (MBAFF) decision.

16. The method according to claim 14 , wherein the super-MB comprises a plurality of pixels, and said classifying step comprises the steps of:

classifying at least some of the plurality of pixels as an edge pixel using an adaptive threshold for the band-pass/high-pass filter;

determining a sum by adding a number of edge pixels; and

comparing the sum to another pre-specified threshold to determine whether the super-MB is classified with respect to the frame mode or the field mode.

17. A method for encoding video data for an image that is divisible into super-macroblocks (super-MBs), comprising:

performing a picture adaptive frame field (PAFF) decision using a band-pass/high-pass filter applied vertically to the super-MBs to respectively classify each of the super-MBs as a moving region or a non-moving region in the image, and wherein the PAFF decision is based on a percentage of super-MBs classified as moving regions in the image.

18. The method according to claim 17 , wherein the band-pass/high-pass filter is an edge detection filter.

19. The method according to claim 17 , wherein the band-pass/high-pass filter uses an adaptable threshold that is selected to distinguish moving edges from true edges.

20. The method according to claim 17 , wherein the PAFF decision is combined with other criteria comprising at least one of frame type, absolute difference of a current picture and previous reference pictures, macroblock characteristics, frame variance, and macroblock variance.

21. The method according to claim 17 , wherein the PAFF decision is performed at a Group Of Pictures (GOPs) level, for all frames in a current GOP, such that the field mode is used for the current GOP when a number of frames decided to be coded in the field mode is larger than a pre-specified percentage, otherwise the frame mode is used for the current GOP.

22. The method according to claim 19 , wherein the threshold is adaptable frame-wise, Group Of Picture (GOP)-wise, or scene-wise.

23. The method according to claim 19 , wherein the band-pass/high-pass filter is applied to one field image, and the adaptable threshold is selected where a number of edge pixels in the one field image is lower than another threshold.

24. The method according to claim 22 , wherein the method further comprises the step of attributing weights to the edge pixels according to at least one of position within a Super-MB and correlation with neighboring pixels prior to thresholding.

25. A method for encoding video data for an image, comprising the step of utilizing a pre-analysis decision whether to classify an individual super-macroblock (super-MB) in the image with respect to a field mode or a frame mode for both a picture adaptive frame field (PAFF) mode selection and a macroblock adaptive frame field (MBAFF) mode selection,

wherein the pre-analysis decision whether to classify an individual super-MB in field mode or frame mode is determined responsive to exceeding a motion percentage threshold for the super-MB.