Identifying an individual for a role

• US 8,718,337 B1
• Filed: 06/30/2010
• Issued: 05/06/2014
• Est. Priority Date: 06/30/2010
• Status: Expired due to Fees

• Current Assignee: IMDb.com, Inc. (Amazon.com, Inc.)
• Original Assignee: IMDb.com, Inc. (Amazon.com, Inc.)
• Inventors: Syrett, Matthew
• Primary Examiner(s): CHAWAN, SHEELA C
• Application Number:
  US12/826,986
• Time in Patent Office: 1,406 Days
• Field of Search: 382/118, 382/190, 382/156, 382/254, 382/291, 382/181, 382/155, 382/276, 382/286, 382/100, 382/192, 382/203, 446/100, 446/372, 446/391, 446/268, 446/337, 446/321, 375/E7.083, 375/E7.076, 30/47, 305/27, 30/50, 305/32, 30/32, 705 1- 2
• US Class Current: 382/118
• CPC Class Codes: G06V 40/172 Classification, e.g. identification

Abstract

Systems and methods are provided for identifying an individual(s) whose face closely resembles a face that is ideal for a role, or who has a facial feature or combination of facial features that closely resembles a facial feature or combination of facial features that is ideal for the role, and who—in addition to having an ideal face, facial feature, or combination of facial features—possesses at least one trait or qualification that is preferable for the role. For example, the systems and methods can be used to identify at least one individual, such as an actor or model, whose face or facial features closely resembles a face or facial features that is ideal for a role, such as a role in a recorded or live performance, a photo shoot, or a fashion show, and who possesses the at least one trait or qualification, such as gender, height, weight, ability to speak a particular language, star power, or a guild affiliation, that is preferable for the role.

Claims

1. A computer-implemented method of identifying at least one actor for a role based on facial resemblance and at least one trait or qualification, the method comprising:

under the control of one or more computer systems configured with executable instructions,

receiving a casting request including at least one image of a target face or target facial feature and one or more traits or qualifications;

processing the image of the target face or target facial feature to extract target facial data that represents the target face or target facial feature, wherein the target facial data is in a digital image format;

analyzing data for each of a plurality of actors to identify actors having the one or more traits or qualifications;

for each identified actor having the one or more traits or qualifications,

extracting facial data from one or more images of the identified actor, the facial data being representative of the face or at least one facial feature of the actor, wherein the facial data is in a digital image format; and

comparing the facial data representative of the face or at least one facial feature of the actor to the target facial data that represents the target face or target facial feature; and

selecting at least one actor in response to the casting request whose facial data most closely matches the target facial data.

2. The method of claim 1, wherein a 2-D based methodology is used to select the at least one actor whose facial data most closely matches the target facial data.

3. The method of claim 2, wherein the 2-D based methodology is a principal component analysis.

4. The method of claim 1, wherein the target face or facial feature is ideal and the at least one trait or qualification is desirable for a role in a photo shoot, a fashion show, or a recorded or live performance.

5. The method of claim 1, wherein the at least one trait or qualification is selected from a group consisting of a gender, an ethnicity, a height, a weight, a body type, a language spoken, a guild affiliation, a number of credits, a country of citizenship, and a country or city of residence.

6. A computer-implemented method of identifying an individual having a face or a facial feature similar to an ideal face or facial feature and possessing at least one qualifying trait or qualification, the method comprising:

under the control of one or more computer systems configured with executable instructions,

receiving an indication of the at least one trait or qualification;

receiving facial data representative of the ideal face or facial feature;

searching at least one data store having facial data and trait or qualification data of each of a plurality of individuals; and

identifying at least one individual whose facial data closely corresponds to the facial data of the ideal face or facial feature and whose trait or qualification data includes the at least one qualifying trait, wherein the facial data is in a digital image format.

7. The method of claim 6, further comprising:

calculating a similarity score for each individual identified as being at least one individual whose facial data closely corresponds to the facial data of the ideal face or facial feature and whose trait or qualification data includes the at least one qualifying trait, the similarity score representing the degree of similarity between the facial data of the identified at least one individual and the facial data of the ideal face or facial feature; and

ranking-ordering the identified individuals based on the similarity scores.

8. The method of claim 6, wherein receiving facial data comprises:

receiving an image of an ideal face or facial feature; and

processing the image to extract facial data representative of the ideal face or facial feature.

9. The method of claim 6, wherein receiving facial data comprises:

receiving a reference to a known individual who has an ideal face or facial feature;

accessing a data store having facial information representative of the face or facial feature of the known individual; and

obtaining facial data representative of the face or facial feature of the known individual.

10. The method of claim 6, wherein identifying the at least one individual whose facial data closely corresponds to the facial data representative of the ideal face or facial feature comprises:

comparing the facial data representative of the ideal face or facial feature to facial data of each of the individuals in the data store.

11. The method of claim 10, wherein the facial data representative of the ideal face or facial feature is a combination of facial data representative of a first facial feature of a first individual and facial data representative of a second facial feature of a second individual.

12. The method of claim 11, where the at least one individual whose facial data closely corresponds to the facial data of the ideal face or facial feature has facial data that corresponds to the combination of facial data representative of the first facial feature of the first individual and facial data representative of the second facial feature of the second individual.

13. A computer-implemented method of identifying an individual whose a face has a combination of ideal facial features and who possesses at least one qualifying trait or qualification, the method comprising:

under the control of one or more computer systems configured with executable instructions,

receiving a reference to a first facial feature of a first known individual and a reference to a second facial feature of a second known individual;

receiving a description of a trait or qualification;

accessing at least one data store to obtain facial data representative of the first facial feature of the first known individual and facial data representative of the second facial feature of the second known individual; and

searching the at least one data store to identify at least one individual whose facial data includes facial data that closely corresponds to the facial data that represents the first facial feature of the first known individual and facial data that closely corresponds to the facial data that represents the second facial feature of the second known individual, wherein the facial data is in a digital image format.

14. The method of claim 13, wherein the first facial feature of the first known individual is a nose.

15. The method of claim 13, wherein the first facial feature of the second known individual is a set of eyes.

16. A computer program product embedded in a non-transitory computer-readable medium for identifying an individual having a face or a facial feature similar to an ideal face or facial feature and possessing at least one qualifying trait or qualification, comprising:

program code for receiving an indication of the at least one trait or qualification;

program code for receiving facial data representative of the ideal face or facial feature;

program code for searching at least one data store having facial data and trait or qualification data of each of a plurality of individuals; and

program code for identifying at least one individual whose facial data closely corresponds to the facial data of the ideal face or facial feature and whose trait or qualification data includes the at least one qualifying trait, wherein the facial data is in a digital image format.

17. The computer program product of claim 16, wherein receiving facial data comprises:

receiving an image of an ideal face or facial feature; and

processing the image to extract facial data representative of the ideal face or facial feature.

18. The computer program product of claim 16, wherein receiving facial data comprises:

receiving a reference to a known individual who has an ideal face or facial feature;

accessing a data store having facial information representative of the face or facial feature of the known individual; and

obtaining facial data representative of the face or facial feature of the known individual.

19. The computer program product of claim 16, wherein identifying the at least one individual whose facial data closely corresponds to the facial data representative of the ideal face or facial feature comprises:

comparing the facial data representative of the ideal face or facial feature to facial data of each of the individuals in the data store.

20. The computer program product of claim 19, wherein the facial data representative of the ideal face or facial feature is a combination of facial data representative of a first facial feature of a first individual and facial data representative of a second facial feature of a second individual.

21. The computer program product of claim 20, where the at least one individual whose facial data closely corresponds to the facial data of the ideal face or facial feature has facial data that corresponds to the combination of facial data representative of the first facial feature of the first individual and facial data representative of the second facial feature of the second individual.

22. A system for identifying an individual having a face or a facial feature similar to an ideal face or facial feature and possessing at least one qualifying trait or qualification, comprising:

a processor; and

a memory device including instructions that, when executed by the processor, cause the processor to;

to receive an indication of the at least one trait or qualification;

to receive facial data representative of the ideal face or facial feature;

to search at least one data store having facial data and trait or qualification data of each of a plurality of individuals; and

to identify at least one individual whose facial data closely corresponds to the facial data of the ideal face or facial feature and whose trait or qualification data includes the at least one qualifying trait, wherein the facial data is in a digital image format.

23. The system of claim 22, wherein the memory device further includes instructions that, when executed by the processor, cause the processor to:

receive an image of an ideal face or facial feature; and

process the image to extract facial data representative of the ideal face or facial feature.

24. The system of claim 22, wherein receiving facial data comprises:

receiving a reference to a known individual who has an ideal face or facial feature;

accessing a data store having facial information representative of the face or facial feature of the known individual; and

obtaining facial data representative of the face or facial feature of the known individual.

25. The system of claim 22, wherein identifying the at least one individual whose facial data closely corresponds to the facial data representative of the ideal face or facial feature comprises:

comparing the facial data representative of the ideal face or facial feature to facial data of each of the individuals in the data store.

26. The system of claim 25, wherein the facial data representative of the ideal face or facial feature is a combination of facial data representative of a first facial feature of a first individual and facial data representative of a second facial feature of a second individual.