Using two-dimensional images of a subject and target models to generate notifications identifying facial feature variation

1 . A method comprising:

receiving, by a processing device, first two-dimensional (2D) image data corresponding to a first 2D image of exterior facial features of a human face of a subject;

identifying, among a plurality of target models, first metrics corresponding to a first target model of a target human face, wherein the first metrics identify a first ratio of a first target exterior facial feature to a second target exterior facial feature of the target human face;

generating, based on the first 2D image data, second metrics that represent one or more exterior facial features of the human face of the subject, wherein the second metrics identify a second ratio of a first exterior facial feature of the one or more exterior facial features to a second exterior facial feature of the one or more exterior facial features of the human face of the subject;

determining variation information identifying one or more differences between the first metrics corresponding to the first target model of the target human face and the second metrics that represent the one or more exterior facial features of the human face of the subject, wherein the variation information identifies a difference between the first ratio and the second ratio; and

providing, to a client device, a notification reflecting the variation information.

2 . The method of claim 1 , wherein the notification identifies user instructions pertaining to an application of a beauty product to decrease the one or more differences.

3 . The method of claim 1 , further comprising:

generating, based on the variation information, a visual indicator that reflects the one or more differences between the first metrics corresponding to the first target model of the target human face and the second metrics that represent the one or more exterior facial features of the human face of the subject; and

generating the notification by compositing the visual indicator with the first 2D image of the human face of the subject.

4 . The method of claim 3 , wherein the first target model is a first target 3D model, the method further comprising:

wherein generating, based on the first 2D image data, the second metrics that represent the one or more exterior facial features of the human face of the subject comprises: generating, using the first 2D image data, a first three-dimensional (3D) model of the human face of the subject, wherein the second metrics correspond to the first 3D model;

wherein generating the visual indicator comprises: generating a modified 3D model of the human face of the subject based on the first 3D model of the human face of the subject and the first target 3D model of the target human face, the modified 3D model comprising a region corresponding to the one or more differences between the first metrics corresponding to the first target 3D model and the second metrics corresponding to the first 3D model; and

wherein generating the notification by compositing the visual indicator on the first 2D image of the human face of the subject comprises: compositing the modified 3D model with the first 2D image of the human face.

5 . The method of claim 4 , wherein generating the first 3D model of the human face of the subject, comprises:

identifying a landmark corresponding to an exterior facial feature of the one or more exterior facial features of the human face of the subject.

6 . The method of claim 5 , wherein generating the modified 3D model of the human face comprises:

determining a difference between the landmark corresponding to the exterior facial feature of the human face of the subject and a target landmark of the first target 3D model of the target human face, wherein the target landmark corresponds to the landmark.

7 . The method of claim 3 , further comprising:

receiving second 2D image data corresponding to a second 2D image of the human face of the subject, the second 2D image of the human face comprising a representation of a beauty product applied at a region of the human face of the subject;

generating, based on the second 2D image data, third metrics that represent the one or more exterior facial features with the applied beauty product;

determining additional variation information identifying one or more additional differences between the first metrics corresponding to the first target model and the third metrics that represent the one or more exterior facial features with the applied beauty product;

generating, based on the additional variation information, an additional visual indicator that reflects the one or more additional differences between the first metrics and the third metrics; and

generating the notification by compositing the additional visual indicator to the second 2D image of the human face of the subject.

8 . The method of claim 1 , wherein identifying, among the plurality of target models, the first metrics corresponding to the first target model of the target human face comprises:

providing an indication of the plurality of target models for user selection at the client device; and

receiving, from the client device, an indication of a user selection of the first target model from the plurality of target models.

9 . The method of claim 1 , wherein generating, based on the first 2D image data, the second metrics that represent the one or more exterior facial features of the human face of the subject, comprises:

providing the first 2D image data as input to a machine learning model; and

obtaining one or more outputs of the machine learning model, the one or more outputs identifying the second metrics.

10 . A system comprising:

a memory device; and

a processing device coupled to the memory device, the processing device to perform operations comprising:

receiving first two-dimensional (2D) image data corresponding to a first 2D image of exterior facial features of a human face of a subject;

identifying, among a plurality of target models, first metrics corresponding to a first target model of a target human face, wherein the first metrics identify a first ratio of a first target exterior facial feature to a second target exterior facial feature of the target human face;

generating, based on the first 2D image data, second metrics that represent one or more exterior facial features of the human face of the subject, wherein the second metrics identify a second ratio of a first exterior facial feature of the one or more exterior facial features to a second exterior facial feature of the one or more exterior facial features of the human face of the subject;

determining variation information identifying one or more differences between the first metrics corresponding to the first target model of the target human face and the second metrics that represent the one or more exterior facial features of the human face of the subject, wherein the variation information identifies a difference between the first ratio and the second ratio; and

providing, to a client device, a notification reflecting the variation information.

11 . The system of claim 10 , wherein the notification identifies user instructions pertaining to an application of a beauty product to decrease the one or more differences.

12 . The system of claim 10 , the operations further comprising:

generating, based on the variation information, a visual indicator that reflects the one or more differences between the first metrics corresponding to the first target model of the target human face and the second metrics that represent the one or more exterior facial features of the human face of the subject; and

generating the notification by compositing the visual indicator with the first 2D image of the human face of the subject.

13 . The system of claim 12 , wherein the first target model is a first target 3D model, the operations further comprising:

wherein generating, based on the first 2D image data, the second metrics that represent the one or more exterior facial features of the human face of the subject comprises: generating, using the first 2D image data, a first three-dimensional (3D) model of the human face of the subject, wherein the second metrics correspond to the first 3D model;

wherein generating the visual indicator comprises: generating a modified 3D model of the human face of the subject based on the first 3D model of the human face of the subject and the first target 3D model of the target human face, the modified 3D model comprising a region corresponding to the one or more differences between the first metrics corresponding to the first target 3D model and the second metrics corresponding to the first 3D model; and

wherein generating the notification by compositing the visual indicator on the first 2D image of the human face of the subject comprises: compositing the modified 3D model with the first 2D image of the human face.

14 . The system of claim 12 , the operations further comprising:

receiving second 2D image data corresponding to a second 2D image of the human face of the subject, the second 2D image of the human face comprising a representation of a beauty product applied at a region of the human face of the subject;

generating, based on the second 2D image data, third metrics that represent the one or more exterior facial features with the applied beauty product;

determining additional variation information identifying one or more additional differences between the first metrics corresponding to the first target model and the third metrics that represent the one or more exterior facial features with the applied beauty product;

generating, based on the additional variation information, an additional visual indicator that reflects the one or more additional differences between the first metrics and the third metrics; and

generating the notification by compositing the additional visual indicator to the second 2D image of the human face of the subject.

15 . A non-transitory computer-readable medium comprising instructions that, when executed by a processing device, cause the processing device to perform operations comprising:

receiving first two-dimensional (2D) image data corresponding to a first 2D image of exterior facial features of a human face of a subject;

identifying, among a plurality of target models, first metrics corresponding to a first target model of a target human face, wherein the first metrics identify a first ratio of a first target exterior facial feature to a second target exterior facial feature of the target human face;

generating, based on the first 2D image data, second metrics that represent one or more exterior facial features of the human face of the subject, wherein the second metrics identify a second ratio of a first exterior facial feature of the one or more exterior facial features to a second exterior facial feature of the one or more exterior facial features of the human face of the subject;

determining variation information identifying one or more differences between the first metrics corresponding to the first target model of the target human face and the second metrics that represent the one or more exterior facial features of the human face of the subject, wherein the variation information identifies a difference between the first ratio and the second ratio; and

providing, to a client device, a notification reflecting the variation information.

16 . The non-transitory computer-readable medium of claim 15 , wherein the notification identifies user instructions pertaining to an application of a beauty product to decrease the one or more differences.

17 . The non-transitory computer-readable medium of claim 15 , further comprising:

generating, based on the variation information, a visual indicator that reflects the one or more differences between the first metrics corresponding to the first target model of the target human face and the second metrics that represent the one or more exterior facial features of the human face of the subject; and

generating the notification by compositing the visual indicator with the first 2D image of the human face of the subject.

18 . The non-transitory computer-readable medium of claim 17 , wherein the first target model is a first target 3D model, the operations further comprising:

wherein generating, based on the first 2D image data, the second metrics that represent the one or more exterior facial features of the human face of the subject comprises: generating, using the first 2D image data, a first three-dimensional (3D) model of the human face of the subject, wherein the second metrics correspond to the first 3D model;

wherein generating the visual indicator comprises: generating a modified 3D model of the human face of the subject based on the first 3D model of the human face of the subject and the first target 3D model of the target human face, the modified 3D model comprising a region corresponding to the one or more differences between the first metrics corresponding to the first target 3D model and the second metrics corresponding to the first 3D model; and

wherein generating the notification by compositing the visual indicator on the first 2D image of the human face of the subject comprises: compositing the modified 3D model with the first 2D image of the human face.

19 . The non-transitory computer-readable medium of claim 17 , further comprising:

receiving second 2D image data corresponding to a second 2D image of the human face of the subject, the second 2D image of the human face comprising a representation of a beauty product applied at a region of the human face of the subject;

generating, based on the second 2D image data, third metrics that represent the one or more exterior facial features with the applied beauty product;

determining additional variation information identifying one or more additional differences between the first metrics corresponding to the first target model and the third metrics that represent the one or more exterior facial features with the applied beauty product;

generating, based on the additional variation information, an additional visual indicator that reflects the one or more additional differences between the first metrics and the third metrics; and

generating the notification by compositing the additional visual indicator to the second 2D image of the human face of the subject.