Demographic analysis of facial landmarks
View Patent ↗A facial image may be annotated with the plurality of facial landmarks. These facial landmarks may be points or regions of the face that are indicative, either alone or in combination with other facial landmarks, of at least one demographic characteristic. Demographic characteristics include, for example, age, race, and/or gender. Based on the demographic characteristic being analyzed, one or more of these facial landmarks may be selected and arranged into an input vector. Then, the input vector may be compared to one or more of the training vectors. An outcome of this comparison may involve in the given facial image being classified into a category germane to the analyzed demographic characteristic (e.g., an age range or age, a racial category, and/or a gender).
1. A method comprising:
obtaining an input vector of facial landmarks, wherein the facial landmarks of the input vector are derived from a given facial image, and wherein the facial landmarks of the input vector are identified based on their ability to represent anthropometric characteristics of the given facial image;
performing, by a computing device, a weighted comparison between the input vector and each of a plurality of training vectors, wherein each training vector is mapped to at least one of a plurality of categories, wherein each training vector represents facial landmarks derived from a different respective facial image, wherein each training vector is associated with an age of an individual from whom the different respective facial image was derived, and wherein each category in the plurality of categories comprises a different age range, wherein at least some of the facial landmarks of the input vector represent aging characteristics of the given facial image, wherein the plurality of categories comprises an adult age range, and wherein at least some of the subset of the training vectors that map to the adult age range represent facial features indicative of wrinkling, hyper-pigmentation, or thinning of fat padding;
based on a result of the weighted comparison, classifying the given facial image into a category of the plurality of categories;
performing a second comparison of the input vector to a subset of the training vectors, wherein at least some of the subset of the training vectors map to the category of the plurality of categories in which the given facial image was classified; and
based on a result of the second comparison, estimating an age of an individual from whom the given facial image was derived.
2. The method of claim 1 , wherein each training vector also maps to a region in an m-dimensional space, wherein there is one region in the m-dimensional space for each category in the plurality, wherein performing the weighted comparison comprises mapping the input vector to at least one of the regions.
3. The method of claim 2 , wherein classifying the given facial image into the category of the plurality comprises selecting the category of the plurality based on the at least one region in the m-dimensional space to which the input vector maps.
4. The method of claim 1 , wherein each category in the plurality of categories comprises a different human racial identity, and wherein at least some of the facial landmarks of the input vector represent racial characteristics of the given facial image.
5. The method of claim 1 , wherein each category in the plurality of categories comprises a different gender, and wherein at least some of the facial landmarks of the input vector represent gender characteristics of the given facial image.
6. The method of claim 1 , wherein the plurality of categories comprises a child age range, and wherein at least some of the subset of the training vectors that map to the child age range represent facial features indicative of bone expansion or hard-tissue elongation.
7. The method of claim 1 , wherein the second comparison comprises fitting the input vector to points in m-dimensional space, wherein the points in the m-dimensional space are defined by the at least some of the training vectors.
8. The method of claim 1 , wherein a mean absolute error of the estimated ages of respective individuals and actual ages of the respective individuals is less than about 4 years.
9. A method comprising:
determining an application to apply to a first input vector of facial landmarks, wherein each facial landmark in the first input vector is derived from a given facial image, and wherein the facial landmarks of the first input vector are identified based on their ability to represent anthropometric characteristics of the given facial image;
based on the determined application, selecting a subset of the facial landmarks from the first input vector to form a second input vector;
comparing, by a computing device, the second input vector to a set of training vectors, wherein each training vector is associated with an age of an individual from whom the different respective facial image was derived; and
based on a result of the comparison, classifying the given facial image into a category of a plurality of categories, wherein the categories in the plurality are defined by the determined application, wherein each category in the plurality of categories comprises a different age range, wherein at least some of the facial landmarks of the input vector represent age characteristics of the given facial image, wherein the plurality of categories comprises an adult age range, and wherein at least some of the subset of the training vectors that map to the adult age range represent facial features indicative of wrinkling, hyper-pigmentation, or thinning of fat padding;
performing a second comparison of the input vector to a subset of the training vectors, wherein at least some of the subset of the training vectors map to the category of the plurality of categories in which the given facial image was classified; and
based on a result of the second comparison, estimating an age of an individual from whom the facial image was derived.
10. The method of claim 9 , wherein each training vector is mapped to at least one category of the plurality, and wherein each training vector represents facial landmarks derived from a different respective facial image.
11. The method of claim 9 , wherein each training vector is mapped to a region in an m-dimensional space, wherein there is one region in the m-dimensional space for each category in the plurality, wherein performing the weighted comparison comprises fitting the input vector to at least one of the regions.
12. The method of claim 9 , wherein classifying the given facial image into the category of the plurality of categories comprises selecting the category of the plurality of categories based on the at least one region in m-dimensional space to which the input vector maps.
13. The method of claim 9 , wherein the plurality of categories comprises a child age range, and wherein at least some of the subset of the training vectors that map to the child age range represent facial features indicative of bone expansion or hard-tissue elongation.
14. An article of manufacture including a non-transitory computer-readable medium, having stored thereon software instructions that, upon execution by a computing device, cause the computing device to perform operations comprising:
identifying an input vector of facial landmarks, wherein the facial landmarks of the input vector are derived from a given facial image, and wherein the facial landmarks of the input vector are identified based on their ability to represent anthropometric characteristics of the given facial image;
performing a weighted comparison between the input vector and each of a plurality of training vectors, wherein each training vector is mapped to at least one of a plurality of categories, wherein each training vector represents facial landmarks derived from a different respective facial image, wherein each training vector is associated with an age of an individual from whom the different respective facial image was derived, and wherein each category in the plurality of categories comprises a different age range, wherein at least some of the facial landmarks of the input vector represent aging characteristics of the given facial image, wherein the plurality of categories comprises an adult age range, and wherein at least some of the subset of the training vectors that map to the adult age range represent facial features indicative of wrinkling, hyper-pigmentation, or thinning of fat padding;
based on a result of the weighted comparison, classifying the given facial image into a category of the plurality;
performing a second comparison of the input vector to a subset of the training vectors, wherein at least some of the subset of the training vectors map to the category of the plurality of categories in which the given facial image was classified; and
based on a result of the second comparison, estimating an age of an individual from whom the given facial image was derived.