CONTENT IDENTIFICATION AND DISTRIBUTION

• US 20130162676A1
• Filed: 12/21/2012
• Published: 06/27/2013
• Est. Priority Date: 12/23/2011
• Status: Active Grant

• Current Assignee: Zappar Ltd.
• Original Assignee: Zappar Ltd.
• Inventors: Taylor, Simon John
• Granted Patent: US 8,814,048 B2
• Time in Patent Office: Days
• US Class Current: 345/633
• CPC Class Codes: G06T 2207/30204 Marker
• CPC Class Codes: G06T 7/75 involving models

Abstract

The invention provides an identifier system for computing identity information from image data. At least part of the image data is representative of an identifier. The identifier comprises a location element and encoded data associated with the location element. The identifier system comprises computer interpretable reference data corresponding to the identifier. The reference data is suitable for use in feature matching to determine a location and an orientation of the location element in the image data—thereby to locate the encoded data in the image data for subsequent decoding into the identity information The invention also provides a computer implemented method of presenting an augmented reality view of a physical article using the identifier system.

Claims

1. An identifier system for computing identity information from image data, at least part of the image data being representative of an identifier, the identifier comprising a location element and encoded data associated with the location element,

the identifier system comprising computer interpretable reference data corresponding to the identifier,

the reference data being suitable for use in feature matching to determine a location and an orientation of the location element in the image data, and thereby to locate the encoded data in the image data for subsequent decoding into the identity information.

2. An identifier system according to claim 1, comprising a candidate feature detector arranged to detect candidate features in the image data, and the feature matching comprising a candidate feature identifier arranged to identify using local context information one or more specific features from the detected candidate features and thereby to determine the location and orientation of the location element.

3. An identifier system according to claim 1, in which:

the reference data comprises a reference image of the location element, and optionally a further reference image of the location element, each reference image being associated with a different combination of scale and orientation of the location element.

4. An identifier system according to claim 1, in which:

the reference data comprises a reference model of at least part of the location element, and optionally a further model of at least part of the location element, each of the reference models being associated with a different combination of scale and orientation of the location element.

5. An identifier system according to claim 3, in which an optimisation algorithm is used to identify a change of scale and/or orientation for fitting the reference image onto the location element of the image data or vice versa.

6. An identifier system according to claim 5, in which:

the reference data comprises a sparse sample of the reference image, or respective sparse samples of the reference images,

each sparse sample being suitable for use, with a corresponding sparse sample of the location element of the image data.

7. An identifier system according to claim 1, in which:

the reference data comprises decoding information which specifies a location relative to the location element from which the encoded data can be obtained.

8. An identifier system according to claim 7, in which:

the decoding information specifies one or more decoding reference values for comparison with the encoded data in order to decode it into the identity information.

9. A computer implemented method of locating an identifier in image data, the identifier comprising a location element and encoded data associated with the location element, the method comprising:

receiving the image data;

receiving computer interpretable reference data corresponding to the identifier;

using feature matching based on the reference data to determine a location and a scale and orientation of the location element in the received image data;

using the determined location, scale and orientation of the location element to locate the encoded data for subsequent decoding into identity information.

10. A method according to claim 9 comprising:

detecting candidate features in the image data; and

identifying using local context information one or more specific features from the detected candidate features to thereby determine the location and orientation of the location element

11. A method according to claim 9, in which the reference data comprises a reference model of at least part of the location element, and in which the feature matching comprises obtaining a model of a part of the received image data and comparing it with the reference model to determine whether it matches the reference model.

12. A method according to claim 11, in which the reference data comprises a further reference model of said at least part of the location element, each of the reference models corresponding to a different combination of scale and orientation of the location element, and in which the feature matching comprises comparing the model of a part of the received image data with the further reference model to determine whether it matches the further reference model.

13. A method according to claim 11, in which the feature matching comprises:

disregarding a potential match in response to determining that the model of a part of the received image data does not match the reference model, or the further reference model.

14. A method according to claim 9, in which the reference data comprises a reference image of the location element, and in which the feature matching comprises comparing a part of the received image data with image data from the reference image to determine whether it matches the reference image.

15. A method according to claim 14, in which the reference data comprises a further reference image of the location element, each of the reference images corresponding to a different combination of scale and orientation of the location element, and in which the feature matching comprises comparing the part of the received image data with image data from the further reference image to determine whether it matches the further reference image.

16. A method according to claim 14,

in which the reference image corresponds to the reference model and the part of the received image data is compared with the image data from the reference image only if the potential match between it and the reference model has not already been disregarded, and

optionally in which the further reference image corresponds to the further reference model and the part of the received image data is compared with the image data from the further reference image only if the potential match between it and the further reference model has not already been disregarded.

17. A method according to claim 11, comprising deriving the location of the location element from the location of the part of the received image data if the model thereof was determined to match the reference model, or the further reference model, and/or if the part of the received image data was determined to match the reference image, or the further reference image.

18. A method according to claim 11, comprising deriving the scale and orientation of the location element from the scale and orientation associated with the reference model, or the further reference model, if the obtained model of the part of the received image data was determined to match the reference model, or the further reference model, respectively.

19. A method according to claim 14, comprising deriving the scale and orientation of the location element from the scale and orientation associated with the reference image, or the further reference image, if the part of the received image data was determined to match the reference image, or the further reference image, respectively.

20. A method according to claim 18, in which determining the scale and orientation of the location element comprises:

using an optimisation algorithm to change the derived scale and/or orientation to fit the location element of the received image data onto a matching reference image, the matching reference image being whichever one of the reference images was determined to match the part of the received image data and/or the reference image corresponding to whichever one of the reference models was determined to match the model of the part of the received image data.

21. A computer implemented method of presenting an augmented reality view of a physical article, comprising:

receiving image data representative of a physical article for which augmented reality content is available;

determining an identity of the augmented reality content;

sending a request for the augmented reality content, the request including an indication of the determined identity;

receiving a response to the request, the response comprising the augmented reality content; and

presenting an augmented reality view of the physical article, based on the image data and the augmented reality content;

in which the physical article includes an identifier, in which the image data comprises data representative of the identifier, and in which determining an identity of the augmented reality content comprises computing identity information from the data representative of the identifier;

in which the identifier comprises a location element and encoded data associated with the location element; and

computing the identity information comprises;

receiving the image data;

receiving computer interpretable reference data corresponding to the identifier;

using feature matching based on the reference data to determine a location and a scale and orientation of the location element in the received image data;

using the determined location, scale and orientation of the location element to locate the encoded data for subsequent decoding into identity information, and

decoding the resulting encoded data into the identity information.