Scene cut detection in a video bitstream

1. A method for detecting scene cuts in a video bitstream comprising the steps of:

determining a number of coded macroblocks for each of a plurality of potential scene cuts;

determining a number of predicted macroblocks for each of the plurality of potential scene cuts;

determining a ratio of the number of coded macroblocks to the number of predicted macroblocks for each of a plurality frames corresponding to the plurality of potential scene cuts;

selecting a frame having a desired ratio;

comparing the ratio to a threshold value; and

detecting a scene cut upon determining that the ratio satisfies the threshold value.

2. The method of claim 1 , further comprising the step of processing a group of frames, wherein each frame includes two fields, and wherein the group of frames comprise a sub-group of pictures and the video bitstream includes a plurality of sub-groups of pictures.

3. The method of claim 1 , further comprising the step of determining whether a first frame is an I or P picture, upon determining that the first frame is not an I or P picture, advancing to a next sub-group of pictures.

4. The method of claim 1 , further comprising the step of determining whether the first picture is an I picture and whether a previous picture was an I or P picture, upon determining the first picture to be a P picture or the previous picture to be a B picture, advancing to a next sub-group of pictures.

5. The method of claim 1 , further comprising the steps of:

initializing the number of predicted macroblocks to zero for each frame; and

initializing the number of coded macroblocks to zero for each frame.

6. The method of claim 1 , wherein at least one of the potential scene cuts exist between interlaced fields of a frame.

7. The method of claim 6 , wherein the step of determining the number of predicted macroblocks further comprises the steps of:

setting the number of predicted macroblocks equal to a number of forward predicted macroblocks for each potential scene cut, upon determining a corresponding field to be a first field in a frame; and

setting the number of predicted macroblocks equal to a number of backward predicted macroblocks for each potential scene cut, upon determining the corresponding field to be a second field in the frame, wherein the frame includes the first and the second fields.

8. The method of claim 6 , wherein the scene cut corresponds to the selected field.

9. The method of claim 1 , wherein the desired ratio is a minimum ratio among all ratios determined.

10. The method of claim 1 , wherein the potential scene cuts exists between the frames of a sub-group of pictures.

11. A method for detecting scene cuts in an MPEG video bitstream comprising the steps of:

determining a number of coded macroblocks for each of a plurality of potential scene cuts, wherein the plurality of potential scene cuts exists between a plurality of frames and between a plurality of fields in each frame;

setting a number of predicted macroblocks equal to a number of forward predicted macroblocks for each of the plurality of potential scene cuts, upon determining a corresponding field to be a first field in a frame, and setting the number of predicted macroblocks equal to a number of backward predicted macroblocks for each of the plurality of potential scene cuts, upon determining the corresponding field to be a second field in the frame, wherein the frame includes the first and the second fields;

determining a ratio of the number of coded macroblocks to the number of predicted macroblocks for each of a plurality fields corresponding to the plurality of potential scene cuts;

selecting a field having a lowest ratio among the plurality of fields;

comparing the lowest ratio to a threshold value; and

detecting a scene cut upon determining that the lowest ratio satisfies the threshold value, wherein the scene cut corresponds to the selected field.

12. The method of claim 11 , further comprising the step of processing a group of frames, each frame including two fields, wherein the group of frames comprise a sub-group of pictures and the video bitstream includes a plurality of sub-groups of pictures.

13. The method of claim 11 , further comprising the steps of:

determining whether a first frame is an I or P picture; and

advancing processing to a next sub-group of pictures, responsive to a determination that the first frame is not an I or P picture.

14. The method of claim 11 , further comprising the steps of:

determining whether the first picture is an I picture;

determining whether a previous picture was an I or P picture; and

advancing processing to a next sub-group of pictures, responsive to a determining the first picture to be a P picture or the previous picture to be a B picture.

15. The method of claim 11 , further comprising the steps of:

initializing the number of predicted macroblocks to zero for each field; and

initializing the number of coded macroblocks to zero for each field.

16. The method of claim 11 , wherein at least one of the plurality of potential scene cuts exist between interlaced fields of a frame.

17. A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps for detecting scene cuts in a video bitstream, the method steps comprising:

determining a number of coded macroblocks for each of a plurality of potential scene cuts;

determining a number of predicted macroblocks for each of the plurality of potential scene cuts;

determining a ratio of the number of coded macroblocks to the number of predicted macroblocks for each of a plurality of frames corresponding to the plurality of potential scene cuts;

selecting a frame having a desired ratio;

comparing the ratio to a threshold value; and

detecting a scene cut upon determining that the ratio satisfies the threshold value.

18. The method of claim 17 , wherein the plurality of potential scene cuts exist between interlaced fields of a frame.

19. The method of claim 18 , wherein the step of determining the number of predicted macroblocks further comprises the steps of:

setting the number of predicted macroblocks equal to a number of forward predicted macroblocks for each potential scene cut, upon determining a corresponding field to be a first field in a frame; and

setting the number of predicted macroblocks equal to a number of backward predicted macroblocks for each potential scene cut, upon determining the corresponding field to be a second field in the frame, wherein the frame includes the first and the second fields.

20. The method of claim 18 , wherein the scene cut corresponds to the selected field.

21. The method of claim 17 , wherein the desired ratio is a minimum ratio among all ratios determined.

22. An apparatus for detecting scene cuts in a video bitstream comprising:

means for determining a number of coded macroblocks for each of a plurality of potential scene cuts;

means for determining a number of predicted macroblocks for each of the plurality of potential scene cuts;

means for determining a ratio of the number of coded macroblocks to the number of predicted macroblocks for each of a plurality frames corresponding to the plurality of potential scene cuts;

means for selecting a frame having a desired ratio;

means for comparing the ratio to a threshold value; and

means for detecting a scene cut upon determining that the ratio satisfies the threshold value.