Automatic appeal measurement method

1. A method for estimating a personal appeal of a candidate individual to an observer, comprising:

a) designating a set of proxy individuals;

b) providing one or more digital images for each proxy individual;

c) determining personal appeal values of each proxy individual to the observer;

d) providing one or more digital images for the candidate individual;

e) using a processor to determine image similarity metrics between the digital images for the candidate individual and the digital images for each proxy individual;

f) using a processor to determine similarity values between the candidate individual and each proxy individual responsive to the determined image similarity metrics;

g) using a processor to estimate the personal appeal of the candidate individual to the observer by determining a weighted combination of the personal appeal values for the proxy individuals, wherein the weighted combination uses weighting coefficients that are determined responsive to the similarity values between the candidate individual and the corresponding proxy individual; and

h) storing an indication of the estimated personal appeal in a processor-accessible memory.

2. The method of claim 1 wherein personal appeal values for at least a subset of the proxy individuals are determined by having the observer evaluate the personal appeal of the proxy individuals.

3. The method of claim 2 wherein the evaluation of the personal appeal of the proxy individuals includes having the observer evaluate the physical attractiveness of the proxy individuals based on viewing the digital images of the proxy individuals.

4. The method of claim 2 wherein the evaluation of the personal appeal of the proxy individuals includes having the observer evaluate non-image information pertaining to the proxy individuals.

5. The method of claim 4 wherein the non-image information includes age, height, weight, hair color, hair style, income, activities, level of education, personality, social affiliations or political affiliations.

6. The method of claim 1 wherein personal appeal values for at least a subset of the proxy individuals are determined by:

(i) designating a set of proxy observers and a set of training individuals, wherein the subset of proxy individuals are included in the set of training individuals;

(ii) determining training personal appeal values of the training individuals to the proxy observers;

(iii) determining observer personal appeal values of at least a subset of the training individuals to the observer;

(iv) determining a proxy observer model that relates the training personal appeal values to the observer personal appeal values; and

(v) using the proxy observer model to determine the personal appeal values of the proxy individuals to the observer responsive to the training personal appeal values.

7. The method of claim 6 wherein the proxy observer model determines the personal appeal value of a particular individual to the observer by determining a weighted combination of corresponding training personal appeal values of the particular individual to the proxy observers, and wherein the weighted combination uses weighting coefficients determined responsive to the training personal appeal values and the observer personal appeal values.

8. The method of claim 7 wherein the weighted combination is a weighted summation, and wherein the weighting coefficients are determined using a linear least squares regression algorithm.

9. The method of claim 1 wherein the image similarity metric is a facial similarity metric.

10. The method of claim 1 wherein the similarity values are also determined responsive to the similarity of non-image information pertaining to the candidate individual and the proxy individuals.

11. The method of claim 10 wherein the non-image information includes age, height, weight, hair color, hair style, income, activities, level of education, personality, social affiliations or political affiliations.

12. The method of claim 1 wherein the weighted combination of the personal appeal values is a weighted summation, and wherein the weighting coefficient for a particular proxy individual is determined by normalizing the similarity values for the particular proxy individual by the sum of the similarity values for all of the proxy individuals.

13. The method of claim 1 further including displaying an indication of the estimated personal appeal to the observer using a user interface on a display device.

14. The method of claim 13 wherein the estimated personal appeal for a plurality of candidate individuals is simultaneously displayed on the user interface.

15. The method of claim 14 wherein the candidate individuals are ordered on the user interface according to the estimated personal appeal.

16. The method of claim 1 further including:

estimating a personal appeal of the observer to the candidate individual; and

determining a mutual personal appeal between the candidate individual and the observer by combining the estimated personal appeal of the candidate individual to the observer with the estimated personal appeal of the observer to the candidate individual.

17. The method of claim 16 wherein the mutual personal appeal is determined by computing the sum or the product of the estimated personal appeal of the candidate individual to the observer and the estimated personal appeal of the observer to the candidate individual.

18. The method of claim 1 wherein estimated personal appeal values are determined for a set of observers, and wherein the estimated personal appeal values are combined to determine a combined appeal value.

19. The method of claim 18 wherein the set of observers are selected to satisfy a set of demographic criteria.

20. The method of claim 19 wherein combined appeal values are determined for a set of candidate individuals who are being considered for use in an advertisement targeted at consumers satisfying the set of demographic criteria.

21. A computer program product for estimating a personal appeal of a candidate individual to an observer comprising an executable software application stored on a non-transitory tangible computer readable storage medium for causing a data processing system to perform the steps of:

a) designating a set of proxy individuals;

b) providing one or more digital images for each proxy individual;

c) determining personal appeal values of each proxy individual to the observer;

d) providing one or more digital images for the candidate individual;

e) determining image similarity metrics between the digital images for the candidate individual and the digital images for each proxy individual;

f) determining similarity values between the candidate individual and each proxy individual responsive to the determined image similarity metrics;

g) estimating the personal appeal of the candidate individual to the observer by determining a weighted combination of the personal appeal values for the proxy individuals, wherein the weighted combination uses weighting coefficients that are determined responsive to the similarity values between the candidate individual and the corresponding proxy individual; and

h) storing an indication of the estimated personal appeal in a processor-accessible memory.