Liveness detection for face capture
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for liveness detection are disclosed. In one aspect, a method includes the actions of providing, for display on a screen to a user, a graphical item that travels along a path. The actions further include tracking a movement of an eye of the user while the graphical item travels along the path on the screen. The actions further include comparing the movement of the eye of the user to the path traveled by the graphical item. The actions further include generating an eye correlation score that reflects a correlation between the movement of the eye of the user and the path traveled by the graphical item. The actions further include determining whether the eye correlation score satisfies an eye liveness threshold score. The actions further include determining whether the user is a live person.
1. A computer-implemented method comprising:
providing, for display on a screen to a user, a graphical item that travels along a path;
based on providing, for display on the screen, the graphical item, activating a light source (i) that projects light in a direction away from the screen and (ii) that is separate from the screen;
receiving, from a camera, video data that includes a representation of the user;
tracking a reflection of the light source in the video data while the graphical item travels along the path on the screen;
comparing the movement of the reflection of the light source to the path traveled by the graphical item;
based on comparing the movement of the reflection of the light source to the path traveled by the graphical item, generating a correlation score that reflects a correlation between the movement of the reflection of the light source and the path traveled by the graphical item;
determining a level of accuracy that is required for determining whether the user is a live person;
based on (i) the level of accuracy that is required for determining whether the user is a live person and (ii) a complexity of the path traveled by the graphical item or whether other objects or instructions where displayed on the screen while the graphical item traveled along the path, selecting, from among multiple liveness threshold scores, a liveness threshold score;
determining whether the correlation score satisfies the liveness threshold score; and
based on determining whether the correlation score satisfies the liveness threshold score, determining whether the user is a live person.
2. The method of claim 1 , wherein:
determining whether the correlation score satisfies the liveness threshold score comprises determining that the correlation score satisfies the liveness threshold score, and determining whether the user is a live person comprises determining that the user is a live person based on determining that the correlation score satisfies the liveness threshold score.
3. The method of claim 1 , wherein:
determining whether the correlation score satisfies the liveness threshold score comprises determining that the correlation score does not satisfy the liveness threshold score, and
determining whether the user is a live person comprises determining that the user is not a live person based on determining that the correlation score does not satisfy the liveness threshold score.
4. The method of claim 1 , path traveled by the graphical item is a random path.
5. The method of claim 1 , comprising:
tracking a movement of a head of the user while the graphical item travels along the path;
comparing the movement of the head of the user to the path traveled by the graphical item;
based on comparing the movement of the head of the user to the path traveled by the graphical item, generating a head correlation score that reflects a correlation between the movement of the head of the user and the path traveled by the graphical item; and
determining whether the head correlation score satisfies a head liveness threshold score,
wherein determining whether the user is a live person is based further on determining whether the head correlation score satisfies the head liveness threshold score.
6. The method of claim 1 , wherein the graphical item travels continuously along the path.
7. The method of claim 1 , wherein the graphical item travels along the path stopping at particular points along the path.
8. The method of claim 1 , wherein an eye of the user reflects the light source.
9. The method of claim 1 , comprising:
receiving, from an application, a request to determine whether the user is a live person; and
based on the application, determining the level of accuracy that is required for determining whether the user is a live person.
10. A system comprising:
one or more computers; and
one or more data storage apparatuses storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
providing, for display on a screen to a user, a graphical item that travels along a path;
based on providing, for display on the screen, the graphical item, activating a light source (i) that projects light in a direction away from the screen and (ii) that is separate from the screen;
receiving, from a camera, video data that includes a representation of the user;
tracking a reflection of the light source in the video data while the graphical item travels along the path on the screen;
comparing the movement of the reflection of the light source to the path traveled by the graphical item;
based on comparing the movement of the reflection of the light source to the path traveled by the graphical item, generating a correlation score that reflects a correlation between the movement of the reflection of the light source and the path traveled by the graphical item;
determining a level of accuracy that is required for determining whether the user is a live person;
based on (i) the level of accuracy that is required for determining whether the user is a live person and (ii) a complexity of the path traveled by the graphical item or whether other objects or instructions where displayed on the screen while the graphical item traveled along the path, selecting, from among multiple liveness threshold scores, a liveness threshold score;
determining whether the correlation score satisfies the liveness threshold score; and
based on determining whether the correlation score satisfies the liveness threshold score, determining whether the user is a live person.
11. The system of claim 10 , wherein:
determining whether the correlation score satisfies the liveness threshold score comprises determining that the correlation score satisfies the liveness threshold score, and
determining whether the user is a live person comprises determining that the user is a live person based on determining that the correlation score satisfies the liveness threshold score.
12. The system of claim 10 , wherein:
determining whether the correlation score satisfies the liveness threshold score comprises determining that the correlation score does not satisfy the liveness threshold score, and
determining whether the user is a live person comprises determining that the user is not a live person based on determining that the correlation score does not satisfy the liveness threshold score.
13. The system of claim 10 , path traveled by the graphical item is a random path.
14. The system of claim 10 , wherein the operations further comprise:
tracking a movement of a head of the user while the graphical item travels along the path;
comparing the movement of the head of the user to the path traveled by the graphical item;
based on comparing the movement of the head of the user to the path traveled by the graphical item, generating a head correlation score that reflects a correlation between the movement of the head of the user and the path traveled by the graphical item; and
determining whether the head correlation score satisfies a head liveness threshold score,
wherein determining whether the user is a live person is based further on determining whether the head correlation score satisfies the head liveness threshold score.
15. The method of claim 10 , wherein the graphical item travels continuously along the path.
16. The system of claim 10 , wherein the graphical item travels along the path stopping at particular points along the path.
17. A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:
providing, for display on a screen to a user, a graphical item that travels along a path;
based on providing, for display on the screen, the graphical item, activating a light source (i) that projects light in a direction away from the screen and (ii) that is separate from the screen;
receiving, from a camera, video data that includes a representation of the user;
tracking a reflection of the light source in the video data while the graphical item travels along the path on the screen;
comparing the movement of the reflection of the light source to the path traveled by the graphical item;
based on comparing the movement of the reflection of the light source to the path traveled by the graphical item, generating a correlation score that reflects a correlation between the movement of the reflection of the light source and the path traveled by the graphical item;
determining a level of accuracy that is required for determining whether the user is a live person;
based on (i) the level of accuracy that is required for determining whether the user is a live person and (ii) a complexity of the path traveled by the graphical item or whether other objects or instructions where displayed on the screen while the graphical item traveled along the path, selecting, from among multiple liveness threshold scores, a liveness threshold score;
determining whether the correlation score satisfies the liveness threshold score; and
based on determining whether the correlation score satisfies the liveness threshold score, determining whether the user is a live person.
18. The medium of claim 17 , wherein:
determining whether the correlation score satisfies the liveness threshold score comprises determining that the correlation score satisfies the liveness threshold score, and
determining whether the user is a live person comprises determining that the user is a live person based on determining that the correlation score satisfies the liveness threshold score.
19. The medium of claim 17 , wherein:
determining whether the correlation score satisfies the liveness threshold score comprises determining that the correlation score does not satisfy the liveness threshold score, and
determining whether the user is a live person comprises determining that the user is not a live person based on determining that the correlation score does not satisfy the liveness threshold score.
20. The medium of claim 17 , wherein the operations further comprise:
tracking a movement of a head of the user while the graphical item travels along the path;
comparing the movement of the head of the user to the path traveled by the graphical item;
based on comparing the movement of the head of the user to the path traveled by the graphical item, generating a head correlation score that reflects a correlation between the movement of the head of the user and the path traveled by the graphical item; and
determining whether the head correlation score satisfies a head liveness threshold score,
wherein determining whether the user is a live person is based further on determining whether the head correlation score satisfies the head liveness threshold score.
21. The medium of claim 17 , wherein the graphical item travels continuously along the path.
22. The medium of claim 17 , wherein the graphical item travels along the path stopping at particular points along the path.