Robust and stealthy video watermarking

• US 20020172394A1
• Filed: 04/24/2001
• Published: 11/21/2002
• Est. Priority Date: 04/24/2001
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Venkatesan, Ramarathnam, Mihcak, M. Kivanc
• Granted Patent: US 6,975,743 B2
• Time in Patent Office: Days
• US Class Current: 382/100
• CPC Class Codes: G06T 1/005 Robust watermarking, e.g. average attack or collusion attack resistant
• CPC Class Codes: G06T 1/0085 Time domain based watermarking, e.g. watermarks spread over several images
• CPC Class Codes: G06T 2201/0065 Extraction of an embedded watermark; Reliable detection
• CPC Class Codes: H04N 19/467 characterised by the embedded information being invisible, e.g. watermarking

Abstract

An implementation of a technology is described herein for the protection of rights in the content of a video sequence. This technology further generally relates to a technology facilitating embedding imperceptible, de-synchronization-resistant watermarks in video sequence and facilitating detecting such watermarks. These watermarks are resistant against unintentional and intentional modifications. In particular, the watermarks are resistant to de-synchronization. In addition, the watermarks are perceptually invisible. The watermarks are hidden in the video so that flicker is minimized or eliminated. More specifically, this technology hides a watermark (of portions thereof) over one or more regions of successive frames. Each region has a center defined by a hash value. A watermark (of portions thereof) is encoded into the region in a “plateau” manner. The mark (of portions thereof) is fully encoded in the frames surrounding the region'"'"'s center, but trail off towards the edges of the region. This abstract itself is not intended to limit the scope of this patent. The scope of the present invention is pointed out in the appending claims.

Claims

1. A method for facilitating the insertion of information into a video signal, the method comprising:

embedding an information pattern in one or more regions of the video signal, a region comprising a plurality successive frames;

wherein, for one or more regions, the information pattern is embedded into a frame of a region with a relative degree of intensity and _relative1oneone one or more frames of that region have the patterns embedded therein which have a relative intensity_relative1that differs from the relative intensity of the pattern embedded in other frames of the region.

2. A method as recited in claim 1 for facilitating the protection of a video signal, wherein the information pattern is a watermark.

3. A method as recited in claim 1 further comprising locating the one or more regions of the video signal for embedding an information pattern therein.

4. A method as recited in claim 1 further comprising randomly locating the one or more regions of the video signal for embedding an information pattern therein.

5. A method as recited in claim 1 further comprising locating the one or more regions of the video signal by randomly selecting frames of the signal, where each of the regions includes at least one of the selected frames.

6. A method as recited in claim 1 further comprising specifying the dimensions of the one or more regions of the video signal.

7. A method as recited in claim 1 further comprising determining hash values of one or more frames of one or more of the regions of the video signal.

8. A method as recited in claim 1 further comprising:

determining hash values of one or more frames of one or more of the regions of the video signal;

persisting the hash values with an association with the video signal.

9. A method as recited in claim 1, wherein, for one or more regions, the information pattern is embedded into a frame of a region in a plateau-shaped manner.

10. A method as recited in claim 1, wherein, for one or more regions, the relative intensity of information pattern embedded into a group of successive frames of the plurality of a region is substantially identical.

11. A method as recited in claim 1, wherein, for one or more regions, the relative intensity of information pattern embedded into a group of successive frames of the plurality of a region is approximately similar.

12. A method as recited in claim 1, wherein, for one or more regions, the relative intensity of information pattern embedded into a group of successive frames of the plurality of a region is gradient.

13. A modulated video signal generated in accordance with a method as recited in claim 1.

14. A computer-readable medium having computer-executable instructions that, when executed by a computer, performs a method as recited in claim 1.

15. A computer comprising one or more computer-readable media having computer-executable instructions that, when executed by the computer, perform a method as recited in claim 1.

17. A method as recited in claim 16, wherein the watermark is embedded into a frame of the region with a relative degree of intensity and _relative1one or more frames of that region have a watermark embedded therein which has a relative intensity_relative1 that differs from the relative intensity of the watermark embedded in other frames of that region.

18. A method as recited in claim 16, wherein the locating comprises randomly locating the one or more regions.

19. A method as recited in claim 16, wherein the locating comprises randomly locating the one or more regions by randomly selecting frames of the signal, where each of the regions includes at least one of the selected frames.

20. A method as recited in claim 16 further comprising specifying the dimensions of the one or more regions of the video signal.

21. A method as recited in claim 16 further comprising determining hash values of one or more frames of one or more of the regions of the video signal.

22. A method as recited in claim 16 fuirther comprising:

determining hash values of one or more frames of one or more of the regions of the video signal;

persisting the hash values with an association with the video signal.

23. A method as recited in claim 16, wherein, for one or more regions, the information pattern is embedded into a frame of a region in a plateau-shaped manner.

24. A method as recited in claim 16, wherein, for one or more regions, the relative intensity of information pattern embedded into a group of successive frames of the plurality of a region is substantially identical

25. A method as recited in claim 16, wherein, for one or more regions, the relative intensity of information pattern embedded into a group of successive frames of the plurality of a region of a region is approximately similar.

26. A method as recited in claim 16, wherein, for one or more regions, the relative intensity of information pattern embedded into a group of successive frames of the plurality of a region is gradient.

27. A modulated video signal generated in accordance with a method as recited in claim 16.

28. A computer-readable medium having computer-executable instructions that, when executed by a computer, performs a method as recited in claim 16.

29. A computer comprising one or more computer-readable media having computer-executable instructions that, when executed by the computer, perform a method as recited in claim 16.

16. A method for facilitating the protection of a video signal, the method comprising:

locating one or more watermark regions of a video signal for embedding a watermark therein, a region comprising a plurality successive frames;

for one or more watermark regions, embedding a watermark in a watermark region in an approximate plateau-shaped manner.

30. A method for facilitating the protection of a video signal, the method comprising:

fully encoding one or more frames of a region of a video signal with a watermark, wherein such frames are fully encoded with the watermark relative to unmarked frames;

partially encoding one or more frames of the region with the watermark, wherein such frames are partially encoded with the watermark relative to the frames of the fully encoding.

31. A modulated video signal generated in accordance with a method as recited in claim 30.

32. A computer-readable medium having computer-executable instructions that, when executed by a computer, performs a method as recited in claim 30.

33. A computer comprising one or more computer-readable media having computer-executable instructions that, when executed by the computer, perform a method as recited in claim 30.

35. A modulated video signal generated in accordance with a method as recited in claim 34.

36. A computer-readable medium having computer-executable instructions that, when executed by a computer, performs a method as recited in claim 34.

37. A computer comprising one or more computer-readable media having computer-executable instructions that, when executed by the computer, perform a method as recited in claim 34.

39. A method as recited in claim 38 further comprising:

examining the one or more detection frames;

determining whether a watermark is present based upon such examining and without access to a known pristine specimen of the video signal.

40. A method as recited in claim 38 further comprising:

examining the one or more detection frames;

determining whether a watermark is present based upon such examining.

41. A method as recited in claim 38 further comprising:

examining the one or more detection frames and frames surrounding one or more detection frames;

determining whether a watermark is present based upon such examining.

42. A method as recited in claim 38, wherein the rep_id of a frame comprise a hash value of the frames.

43. A method as recited in claim 38, wherein the set of one or more expected rep_ids comprise a hash value of frame of an original signal within which a watermark is embedded.

44. A computer-readable medium having computer-executable instructions that, when executed by a computer, performs a method as recited in claim 38.

45. A computer comprising one or more computer-readable media having computer-executable instructions that, when executed by the computer, perform a method as recited in claim 38.

47. A system as recited in claim 46, wherein the information pattern is a watermark.

34. A method for facilitating the protection of a video signal, the method comprising:

fully encoding one or more frames of a region of a video signal with a watermark, wherein the frames are fully encoded with the watermark relative to unmarked frames;

gradiently encoding at least one plurality of successive frames of the region with the watermark, wherein the plurality is gradiently encoded with the watermark relative to the unmarked frames and the fully encoded frames of the fully encoding.

38. A method for facilitating the protection of a video signal, the method comprising:

generating one or more frames of a subject video signal to produce a representative identification (“

rep_id”

) for such frames, wherein a rep_id of a frame is based upon inherent characteristics of such frame;

comparing the representative identification for the frames to a set of one or more expected rep_ids;

selecting one or more detection frames of the subject video signal, wherein the generated rep_id of a detection frame substantially correspond to the an expected rep_id of the set.

46. An information pattern encoding system, the system comprising:

a video signal obtainer configured to obtain a video signal;

a region locator configured to locate one or more regions of the video signal, a region comprising a plurality successive frames;

a region marker configured to embed an information pattern in one or more regions of the video signal, a region comprising a plurality successive frames, wherein, for one or more regions, the information pattern is embedded into a frame of a region with a relative degree of intensity and relative1one oneone or more frames of that region have the patterns embedded therein which have a relative intensity relative1 that differs from the relative intensity of the pattern embedded in other frames of the region.

48. A watermark detecting system, the system comprising:

a video signal obtainer configured to obtain a subject video signal;

a frame locator configured to;

generate one or more frames of the subject video signal to produce a representative identification (“

rep_id”

) for such frames, wherein a rep_id of a frame is based upon inherent characteristics of such frame;

compare the representative identification for the frames to a set of one or more expected rep_ids;

select one or more detection frames of the subject video signal, wherein the generated rep_id of a detection frame substantially correspond to the an expected rep_id of the set;

a watermark detector configured to;

examine the one or more detection frames;

determine whether a watermark is present based upon such examining and without access to a known pristine specimen of the video signal.