Face template balancing
Implementations generally relate to face template balancing. In some implementations, a method includes generating face templates corresponding to respective images. The method also includes matching the images to a user based on the face templates. The method also includes receiving a determination that one or more matched images are mismatched images. The method also includes flagging one or more face templates corresponding to the one or more mismatched images as negative face templates.
1. A computer-implemented method comprising:
providing a target image including a target user;
retrieving from a storage device, at least one positive template associated with a first weight and generated from a set of images previously verified to depict a candidate user, wherein the at least one positive template includes an indication of at least one candidate characteristic of the candidate user;
retrieving from the storage device, one or more negative templates associated with a second weight and generated from a second set of images previously verified to depict one or more other users and to not depict the candidate user, wherein the one or more negative templates include an indication of at least one differentiating characteristic of the one or more other users;
determining whether the target user is identifiable as the candidate user, based, at least in part, on comparing the first weight of the at least one positive template and the second weight of the negative templates; and
in response to determining the target user is identifiable as the candidate user, transmitting the target image and a tag identifying the target user as the candidate user to a computing device through a data communication network.
2. The method of claim 1 , wherein the at least one candidate characteristic includes at least one of an article of clothing, a mark, and an accessory.
3. The method of claim 1 , wherein the at least one differentiating characteristic includes at least one of an article of clothing, a mark, and an accessory.
4. The method of claim 1 , wherein the at least one candidate characteristic includes at least one of: a detail of a characteristic of the candidate user, a location of a characteristic of the candidate user, and a distance between two or more characteristics of the candidate user.
5. The method of claim 1 , wherein the at least one differentiating characteristic includes at least one of: a detail of a characteristic of the candidate user, a location of a characteristic of the candidate user, and a distance between two or more characteristics of the candidate user.
6. The method of claim 1 , wherein the at least one positive template includes two or more positive templates, and wherein the first weight is associated with a difference between the two or more positive templates.
7. The method of claim 1 , wherein the first weight is a positive weight and the second weight is a negative weight.
8. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
providing a target image including a target user;
retrieving from a storage device, at least one positive template associated with a first weight and generated from a set of images previously verified to depict a candidate user, wherein the at least one positive template includes an indication of at least one candidate characteristic of the candidate user;
retrieving from the storage device, one or more negative templates associated with a second weight and generated from a second set of images previously verified to depict one or more other users and to not depict the candidate user, wherein the one or more negative templates include an indication of at least one differentiating characteristic of the one or more other users, wherein the at least one differentiating characteristic and at least one candidate characteristic are different;
determining whether the target user is identifiable as the candidate user, based, at least in part, on comparing the first weight of the at least one positive template and the second weight of the negative templates; and
in response to determining the target user is identifiable as the candidate user, transmitting the target image and a tag identifying the target user as the candidate user to a computing device through a data communication network.
9. The computer-readable medium of claim 8 , wherein the at least one candidate characteristic includes at least one of an article of clothing, a mark, and an accessory.
10. The computer-readable medium of claim 8 , wherein the at least one differentiating characteristic includes at least one of an article of clothing, a mark, and an accessory.
11. The computer-readable medium of claim 8 , wherein the at least one candidate characteristic includes at least one of: a detail of a characteristic of the candidate user, a location of a characteristic of the candidate user, and a distance between two or more characteristics of the candidate user.
12. The computer-readable medium of claim 8 , wherein the at least one differentiating characteristic includes at least one of: a detail of a characteristic of the candidate user, a location of a characteristic of the candidate user, and a distance between two or more characteristics of the candidate user.
13. The computer-readable medium of claim 8 , wherein the at least one positive template includes two or more positive templates, and wherein the first weight is associated with a difference between the two or more positive templates.
14. The computer-readable medium of claim 8 , wherein the first weight is a positive weight and the second weight is a negative weight.
15. A system comprising:
one or more processors; and
one or more computer-readable media having instructions stored thereon that, when executed by the one or more processors, cause performance of operations comprising:
providing a target image including a target user;
retrieving from a storage device, at least one positive template associated with a first weight and generated from a set of images previously verified to depict a candidate user, wherein the at least one positive template includes an indication of at least one candidate characteristic of the candidate user;
retrieving from the storage device, one or more negative templates associated with a second weight and generated from a second set of images previously verified to depict one or more other users and to not depict the candidate user, wherein the one or more negative templates include an indication of at least one differentiating characteristic of the one or more other users;
determining whether the target user is identifiable as the candidate user, based, at least in part, on comparing the first weight of the at least one positive template and the second weight of the negative templates; and
in response to determining the target user is identifiable as the candidate user, transmitting the target image and a tag identifying the target user as the candidate user to a computing device through a data communication network.
16. The system of claim 15 , wherein the at least one candidate characteristic includes at least one of an article of clothing, a mark, and an accessory.
17. The system of claim 15 , wherein the at least one candidate characteristic includes at least one of: a detail of a characteristic of the candidate user, a location of a characteristic of the candidate user, and a distance between two or more characteristics of the candidate user.
18. The system of claim 15 , wherein the at least one differentiating characteristic includes at least one of: a detail of a characteristic of the candidate user, a location of a characteristic of the candidate user, and a distance between two or more characteristics of the candidate user.
19. The system of claim 15 , wherein the at least one positive template includes two or more positive templates, and wherein the first weight is associated with a difference between the two or more positive templates.
20. The system of claim 15 , wherein the first weight is a positive weight and the second weight is a negative weight.