Method and device for temporal segmentation of a video sequence

1. A method of temporal segmentation of a video sequence, the method comprising:

an analysis step, of analyzing one or more markers embedded within the video sequence, wherein content of the one or more markers allows the video sequence to be segmented into a plurality of video sub-sequences.

2. A method according to claim 1 , wherein the method further comprises:

a division step, of dividing the video sequence into a plurality of groups of pictures;

a selection step, of successively selecting each group of the plurality of groups of pictures;

an extraction step, of extracting a binary message from each selected group of pictures;

a validity analysis step, of successively analyzing a validity of each binary message extracted in the extraction step, so as to determine whether the extracted binary message is a marker or a sequence of natural noise; and

a group step, of grouping together groups of pictures according to a message content thereof.

3. A method according to claim 2 , wherein, in the extraction step, a local marker and a global marker are predicted and a first message is extracted from the global marker and a second message is extracted from the local marker.

4. A method according to claim 2 , wherein, in the extraction step, a message size is calculated and the binary message is then extracted based on the message size.

5. A method according to claim 2 , further comprising a video synchronization step, of performing video synchronization.

6. A method according to claim 2 , wherein, in the validity analysis step, a calculation is made of a measure of confidence in a hypothesis that the binary message comes from a group of pictures containing a marker.

7. A method according to claim 2 , wherein, in the validity analysis step, a determination is made as to whether a message extracted from a current group of pictures is identical to a message extracted from a preceding group of pictures.

8. A method according to claim 7 , wherein, in making the determination, a calculation is made of a percentage (P) of bits that is different between the message extracted from the current group of pictures and the message extracted from the preceding group of pictures, and a decision is made that the message extracted from the current group of pictures and the message extracted from the preceding group of pictures are identical if the percentage (P) is less than a predetermined first value.

9. A method according to claim 7 , wherein, in making the determination, a calculation is made of a percentage (P) of bits that is different between the message extracted from the current group of pictures and the message extracted from the preceding group of pictures, and a decision is made that the message extracted from the current group of pictures and the message extracted from the preceding group of pictures are identical if the percentage (P) is less than a predetermined first value and if no temporal de-synchronization is noted or if the percentage (P) is greater than a second predetermined value and if temporal de-synchronization is noted.

10. A method according to claim 2 , wherein, in the validity analysis step, a statistic of errors between a reference message and local messages is studied, and a status and a message of a group of pictures initially detected as not comprising any marker is modified, if a percentage of errors between the local messages and the reference message is within a predetermined criterion of the statistic.

11. A method according to claim 10 , wherein, in the validity analysis step, an average number of erroneous bits between the reference message and the local messages is calculated.

12. A method according to claim 10 or claim 11 , wherein the reference message is calculated from a global group of pictures preceding a local group of pictures for which a change of message has been detected.

13. A method according to claim 11 , further comprising:

a comparison step, of comparing the average number of erroneous bits with a binary error rate calculated beforehand; and,

if the average number of erroneous bits is less than a weighted binary error rate by a predetermined value, a status of the corresponding group of pictures is modified so that, if the corresponding group of pictures was considered as not comprising any marker, the corresponding group of pictures will be considered after modification as comprising a marker.

14. A method according to claim 2 , wherein, in the validity analysis step, statuses and messages linked to local groups of pictures are improved by a feedback loop.

15. A method according to claim 2 , wherein each group of pictures includes 24 consecutive pictures.

16. A digital signal processing apparatus, comprising a device adapted to implement a method according to claim 1 .

17. A device for temporal segmentation of a video sequence comprising:

analysis means for analyzing one or more markers embedded within the video sequence, wherein content of the one or more markers allows the video sequence to be segmented into a plurality of video sub-sequences.

18. A device according to claim 17 , further comprising:

division means for dividing the video sequence into a plurality of groups of pictures;

selection means for successively selecting each group of the plurality of groups of pictures;

extraction means for extracting a binary message from each selected group of pictures;

validity analysis means for successively analyzing a validity of each binary message extracted by the extraction means, so as to determine whether the extracted binary message is a marker or a sequence of natural noise; and

group means for grouping together groups of pictures according to a message content thereof.

19. A device according to claim 18 , wherein the extraction means is adapted to predict a local marker and a global marker and to extract a first message from the global marker and a second message from the local marker.

20. A device according to claim 18 , wherein the extraction means is adapted to calculate a message size and then to extract the binary message based on the message size.

21. A device according to claim 18 , further comprising video synchronization means for performing video synchronization.

22. A device according to claim 18 , wherein the validity analysis means is adapted to calculate a measure of confidence in a hypothesis that the binary message comes from a group of pictures containing a marker.

23. A device according to claim 18 , wherein the validity analysis means comprises test means for testing whether a message extracted from a current group of pictures is identical to a message extracted from a preceding group of pictures.

24. A device according to claim 23 , wherein the test means is adapted to calculate a percentage (P) of bits that is different between the message extracted from the current group of pictures and the message extracted from the preceding group of pictures and to decide that the message extracted from the current group of pictures and the message extracted from the preceding group of pictures are identical if the percentage (P) is less than a first predetermined value.

25. A device according to claim 23 , wherein the test means is adapted to calculate a percentage (P) of bits that is different between the message extracted from the current group of pictures and the message extracted from the preceding group of pictures and to decide that the message extracted from the current group of pictures and the message extracted from the preceding group of pictures are identical if the percentage (P) is less than a first predetermined value and if no temporal de-synchronization is noted or if the percentage (P) is greater than a second predetermined value and if temporal de-synchronization is noted.

26. A device according to claim 18 , wherein the validity analysis means comprises studying means for studying a statistic of errors between a reference message and local messages, and modification means for modifying a status and a message of a group of pictures initially detected as not comprising any marker, if a percentage of errors between the local messages and the reference message is within a predetermined criterion of the statistic.

27. A device according to claim 26 , wherein the validity analysis means comprises means for calculating an average number of erroneous bits between the reference message and the local messages.

28. A device according to claim 26 or claim 27 , wherein the reference message is calculated from a global group of pictures preceding a local group of pictures for which a change of message has been detected.

29. A device according to claim 27 , further comprising:

comparison means for comparing the average number of erroneous bits with a binary error rate calculated beforehand; and

modification means for modifying, if the average number of erroneous bits is less than a weighted binary error rate by a predetermined value, a status of a corresponding group of pictures, so that, if the corresponding group of pictures was considered as not comprising any marker, the corresponding group of pictures will be considered after modification as comprising a marker.

30. A device according to claim 18 , wherein the validity analysis means implements a feedback loop adapted to improve statuses and messages linked to local groups of pictures.

31. A device according to claim 18 , wherein each group of pictures includes 24 consecutive pictures.

32. A digital signal processing apparatus, comprising a device according to claim 17 .