Watermark extractor enhancements based on payload ranking

• US 9,251,549 B2
• Filed: 07/23/2013
• Issued: 02/02/2016
• Est. Priority Date: 07/23/2013
• Status: Active Grant

• Current Assignee: Verance Corporation
• Original Assignee: Verance Corporation
• Inventors: Petrovic, Rade, Winograd, Joseph M., Downes, Patrick George
• Primary Examiner(s): Bali, Vikkram
• Assistant Examiner(s): Mangialaschi, Tracy
• Application Number:
  US13/948,618
• Publication Number:
  US 20150030200A1
• Time in Patent Office: 924 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06T 1/0028 Adaptive watermarking, e.g. Human Visual System [HVS]-based watermarking
• CPC Class Codes: G06T 1/005 Robust watermarking, e.g. average attack or collusion attack resistant
• CPC Class Codes: G06T 2201/0065 Extraction of an embedded watermark; Reliable detection
• CPC Class Codes: G06T 2201/0202 whereby the quality of watermarked images is measured; Measuring quality or performance of watermarking methods; Balancing between quality and robustness

Abstract

Methods, devices, systems and computer program products improve the detection of watermarks that are embedded in a host content by providing information indicative of a ranking of watermark payloads to watermark extractors. A watermark extractor device obtains information indicative of a ranking of a plurality of watermark payload values, where such information designates each watermark payload value within a first subset of the plurality of watermark payload values as having an associated rank value. Based at least in-part on the information indicative of the ranking, the watermark extractor device is configured to prioritize resource allocation for extraction of payload values that are designated as belonging to the first subset over payload values that are not designated as belonging to the first subset. The watermark extractor then extracts a payload value, designated as belonging to the first subset, from one or more watermarks embedded in the host content.

Claims

1. A method, comprising:

receiving a host content at a watermark extractor device;

obtaining, at the watermark extractor device, information indicative of a ranking of a plurality of watermark payload values, the information designating each watermark payload value within a first subset of the plurality of watermark payload values as having an associated rank value;

based at least in-part on the information indicative of the ranking, configuring the watermark extractor device to prioritize resource allocation for extraction of payload values that are designated as belonging to the first subset over payload values that are not designated as belonging to the first subset; and

extracting a payload value from one or more watermarks embedded in the host content,

wherein extracting the payload value comprises;

comparing a candidate watermark payload value obtained from the host content to a plurality of constructed error correction code watermark payload templates corresponding to each of the watermark payload values in the first subset,

producing a plurality of error counts, each error count representing a count of mismatched symbols between the candidate watermark payload value and a corresponding constructed error correction code watermark payload template, and

identifying, as the extracted watermark payload value, a watermark payload value that corresponds to the smallest error count, and

wherein the extracted payload value is one of the payload values designated as belonging to the first subset.

2. The method of claim 1, wherein configuring the watermark extractor device to prioritize resource allocation enables extraction of at least one payload value within the first subset with a higher robustness level compared to extraction of any payload value that is not in the first subset.

3. The method of claim 1, wherein configuring the watermark extractor device to prioritize resource allocation enables extraction of each of the payload values within the first subset with a higher robustness level compared to extraction of each of the payload values within the first subset without the additionally allocated resources.

4. The method of claim 1, wherein the ranking comprises a listing of watermark payload values in a particular order of importance.

5. The method of claim 1, wherein the designation as belonging to the first subset is based at least in-part on having a higher likelihood of being extracted compared to watermark payload values that are not in the first subset.

6. The method of claim 5, wherein the likelihood of being extracted for at least one of the watermark payload values within the first subset is determined based upon a likelihood of being extracted by other watermark extractor devices.

7. The method of claim 1, wherein the designation as belonging to the first subset is based at least in-part on one or more characteristics of the host content.

8. The method of claim 7, wherein the one or more characteristics comprise one or both of:

an age or popularity of the host content.

9. The method of claim 1, wherein the designation as belonging to the first subset is based at least in-part on an extent of economic loss associated with unauthorized dissemination of the host content.

10. The method of claim 1, wherein the designation as belonging to the first subset is based at least in-part on statistics collected regarding one or both of:

the host content, or

one or more watermark payload values.

11. The method of claim 1, wherein the designation as belonging to the first subset is based at least in-part on one or more of:

a viewing history of a user of the extractor device,

a location of a user of the extractor device,

a preference of a user of the extractor device,

a demographic information of a user of the extractor device, or

a behavioral information of a user of the extractor device.

12. The method of claim 1, wherein a count of watermark payload values in the first subset is determined based on a dynamic threshold value that is changeable, thereby allowing designation of a different number of watermark payload values as belonging to the first subset.

13. The method of claim 1, further comprising updating the first subset to include at least one watermark payload value that is different from the watermark payload values before the updating.

14. The method of claim 1, wherein configuring the watermark extractor device to prioritize resource allocation includes one or more of:

prioritizing usage of a memory resource,

prioritizing usage of processing operations,

allocating additional watermark payload templates, or

allocating additional error correction code templates.

15. The method of claim 1, wherein configuring the watermark extractor device to prioritize resource allocation enables extraction of at least one watermark payload value that is designated as belonging to the first subset with an increased robustness level and with an increased probability of false watermark extractions compared to a watermark payload value that is not designated as belonging to the first subset.

16. The method of claim 15, wherein an expected probability of false watermark extractions for all watermark payload values is not increased.

17. The method of claim 1, further comprising allocating additional watermark templates associated with the watermark payload values in the first subset to produce an expanded set of watermark templates, wherein

extracting the payload value comprises comparing the candidate watermark payload value obtained from the host content to all templates in the expanded set of watermark templates to obtain a match to within a predefined error tolerance.

18. The method of claim 17, wherein at least a subset of the expanded set of watermark templates is stored as a look up table (LUT) within the extractor device prior to initiating extraction of the payload value.

19. The method of claim 17, wherein at least a subset of the expanded set of watermark templates is generated on-the-fly, subsequent to initiating extraction of the payload value.

20. The method of claim 1, wherein configuring the watermark extractor device to prioritize resource allocation for extraction of payload values that are designated as belonging to the first subset comprises:

configuring the watermark extractor device to conduct a search for payload values that are not designated as belonging to the first subset using non-template-matching error correction code decoding operations.

21. The method of claim 1, further comprising configuring the watermark extractor device to allocate fewer resources for extraction of watermark payload values that are not designated as belonging to the first subset compared to the watermark payload values that are designated as belonging to the first subset.

22. The method of claim 21, wherein allocated resources for extraction of all payload values is not increased.

23. The method of claim 1, wherein the information indicative of the ranking is obtained from a database located remotely from the extractor device.

24. The method of claim 1, wherein:

each of the watermarked payload values is ranked in an order of importance; and

configuring the watermark extractor device to prioritize resource allocation comprises allocating more resources for extraction of a payload value that is ranked as being more important within the first subset compared to a payload value that is ranked as being less important within the first subset.

25. The method of claim 1, wherein the watermark payload value that corresponds to the smallest error count is identified as the extracted watermark payload value upon a determination that the smallest error count is below a predefined error tolerance value.

26. A device, comprising:

a processor; and

a memory comprising processor executable code, the processor executable code, when executed by the processor, configures the device to;

receive a host content;

obtain information indicative of a ranking of a plurality of watermark payload values, the information designating each watermark payload value within a first subset of the plurality of watermark payload values as having an associated rank value;

based at least in-part on the information indicative of the ranking, configure the device to prioritize resource allocation for extraction of payload values that are designated as belonging to the first subset over payload values that are not designated as belonging to the first subset; and

extract a payload value from one or more watermarks embedded in the host content, wherein extraction of the payload value comprises;

comparison of a candidate watermark payload value obtained from the host content to a plurality of constructed error correction code watermark payload templates corresponding to each of the watermark payload values in the first subset,

production of a plurality of error counts, each error count representing count of mismatched symbols between the candidate watermark payload value and a corresponding constructed error correction code watermark payload template, and

identification, as the extracted watermark payload value, a watermark payload value that corresponds to the smallest error count, and

wherein the extracted payload value is one of the payload values designated as belonging to the first subset.

27. The device of claim 26, wherein configuring the watermark extractor device to prioritize resource allocation enables extraction of at least one payload value within the first subset with a higher robustness level compared to extraction of any payload value that is not in the first subset.

28. The device of claim 26, wherein configuring the watermark extractor device to prioritize resource allocation enables extraction of each of the payload values within the first subset with a higher robustness level compared to extraction of each of the payload values within the first subset without the additionally allocated resources.

29. The device of claim 26, wherein the ranking comprises a listing of watermark payload values in a particular order of importance.

30. The device of claim 26, wherein the designation as belonging to the first subset is based at least in-part on having a higher a likelihood of being extracted compared to watermark payload values that are not in the first subset.

31. The device of claim 30, wherein the likelihood of being extracted for at least one of the watermark payload values within the first subset is determined based upon a likelihood of being extracted by other devices.

32. The device of claim 26, wherein the designation as belonging to the first subset is based at least in-part on one or more characteristics of the host content.

33. The device of claim 32, wherein the one or more characteristics comprise one or both of:

an age or popularity of the host content.

34. The device of claim 26, wherein the designation as belonging to the first subset is based at least in-part on an extent of economic loss associated with unauthorized dissemination of the host content.

35. The device of claim 26, wherein the designation as belonging to the first subset is based at least in-part on statistics collected regarding one or both of:

the host content, or

one or more watermark payload values.

36. The device of claim 26, wherein the designation as belonging to the first subset is based at least in-part on one or more of:

a viewing history of a user of the device,

a location of a user of the device,

a preference of a user of the device,

a demographic information of a user of the extractor device, or

a behavioral information of a user of the extractor device.

37. The device of claim 26, wherein the processor executable code, when executed by the processor, further configures the device to:

receive new information indicative of the ranking; and

modify the first subset in accordance with the new information.

38. The device of claim 26, wherein the processor executable code, when executed by the processor, configures the device to modify the first subset by at least one of:

adding a payload value to the first subset that was not previously in the first subset, or

removing an existing payload value from the first subset.

39. The device of claim 26, the processor executable code, when executed by the processor, configures the device to prioritize resource allocation by including one or more of:

prioritizing usage of a memory resource,

prioritizing usage of processing operations,

allocating additional watermark payload templates, or

allocating additional error correction code templates.

40. The device of claim 26, wherein configuring the device to prioritize resource allocation enables extraction of at least one watermark payload value that is designated as belonging to the first subset with an increased robustness level and with an increased probability of false watermark extractions compared to a watermark payload value that is not designated as belonging to the first subset.

41. The device of claim 40, wherein an expected probability of false watermark extractions for all watermark payload values remains unchanged.

42. The device of claim 26, wherein:

the processor executable code, when executed by the processor, further configures the device to allocate additional watermark templates associated with the watermark payload values in the first subset to produce an expanded set of watermark templates; and

the processor executable code, when executed by the processor, configures the device to extract the payload value by at least in-part comparing the candidate watermark payload value obtained from the host content to all templates in the expanded set of watermark templates to obtain a match to within a predefined error tolerance.

43. The device of claim 42, wherein the processor executable code, when executed by the processor, further configures the device to store at least a subset of the expanded set of watermark templates as a look up table (LUT) within the device prior to initiating extraction of the payload value.

44. The device of claim 42, wherein the processor executable code, when executed by the processor, further configures the device to generate at least a subset of the expanded set of watermark templates on-the-fly, subsequent to initiating extraction of the payload value.

45. The device of claim 26, the processor executable code, when executed by the processor, configures the device to conduct a search for payload values that are not designated as belonging to the first subset using non-template-matching error correction code decoding operations.

46. The device of claim 26, wherein the processor executable code, when executed by the processor, further configures the device to allocate fewer resources for extraction of watermark payload values that are not designated as belonging to the first subset compared to the watermark payload values that are designated as belonging to the first subset.

47. The device of claim 46, wherein allocated resources for extraction of all payload values is not increased.

48. The device of claim 26, wherein:

each of the watermarked payload values is ranked in an order of importance; and

the processor executable code, when executed by the processor, configures the device to prioritize resource allocation by allocating more resources for extraction of a payload value that is ranked as being more important within the first subset compared to a payload value that is ranked as being less important within the first subset.

49. A system comprising the device of claim 26, and further comprising a management center comprising a database, wherein the management center is configured to provide the information indicative of the ranking from the database to the extractor device.

50. The system of claim 48, further comprising a plurality of additional devices, each of the plurality of additional devices comprising a corresponding watermark extractor device, wherein the management center is configured to communicate bi-directionally with each of the plurality of additional devices and with the device.

51. The device of claim 26, wherein the watermark payload value that corresponds to the smallest error count is identified as the extracted watermark payload value upon a determination that the smallest error count is below a predefined error tolerance value.

52. A computer program product, embodied on one or more non-transitory computer readable media, comprising:

program code for receiving a host content at a watermark extractor device;

program code for obtaining, at the watermark extractor device, information indicative of a ranking of a plurality of watermark payload values, the information designating each watermark payload value within a first subset of the plurality of watermark payload values as having an associated rank value;

program code for, based at least in-part on the information indicative of the ranking,

configuring the watermark extractor device to prioritize resource allocation for extraction of payload values that are designated as belonging to the first subset over payload values that are not designated as belonging to the first subset; and

program code for extracting a payload value from one or more watermarks embedded in the host content,

wherein extracting the payload value comprises;

comparing a candidate watermark payload value obtained from the host content to a plurality of constructed error correction code watermark payload templates corresponding to each of the watermark payload values in the first subset,

producing a plurality of error counts, each error count representing a count of mismatched symbols between the candidate watermark payload value and a corresponding constructed error correction code watermark payload template, and

identifying, as the extracted watermark payload value, a watermark payload value that corresponds to the smallest error count, and

wherein the extracted payload value is one of the payload values designated as belonging to the first subset.

53. The A computer program product of claim 52, wherein the watermark payload value that corresponds to the smallest error count is identified as the extracted watermark payload value upon a determination that the smallest error count is below a predefined error tolerance value.

54. A device, comprising:

a receiver configured to receive a host content;

a processor implemented at least in-part in hardware and configured to obtain information indicative of a ranking of a plurality of watermark payload values, the information designating each watermark payload value within a first subset of the plurality of watermark payload values as having an associated value;

the processor further configured to, based at least in-part on the information indicative of the ranking, configure a watermark extractor component to prioritize resource allocation for extraction of payload values that are designated as belonging to the first subset over payload values that are not designated as belonging to the first subset, wherein;

the watermark extractor is configured to extract at least one payload value designated as belonging to the first subset from one or more watermarks embedded in the host content, wherein extraction of the payload value comprises;

comparison of a candidate watermark payload value obtained from the host content to a plurality of constructed error correction code watermark payload templates corresponding to each of the watermark payload values in the first subset,

production of a plurality of error counts, each error count representing a count of mismatched symbols between the candidate watermark payload value and a corresponding constructed error correction code watermark payload template, and

identification, as the extracted watermark payload value, a watermark payload value that corresponds to the smallest error count.

55. The device of claim 54, wherein the watermark payload value that corresponds to the smallest error count is identified as the extracted watermark payload value upon a determination that the smallest error count is below a predefined error tolerance value.