Illumination-based user authentication

1. An apparatus for illumination-based user authentication, the apparatus comprising at least one processor and at least one memory including program code, the at least one memory and the program code configured to, with the processor, cause the apparatus to at least:

receive a user authentication request from a user for a computing device,

and in response to receiving the user authentication request,

cause a display device of the computing device to display a display pattern comprising a series of displayed images during a first time period comprising a plurality of temporal instances,

wherein for each of the plurality of temporal in stances, a different one of the series of displayed images is displayed on the display device,

wherein each of the different one of the series of displayed images corresponds to a defined combination of Red-Green-Blue (RGB) color and position on the display,

determine, based on image sensor data received from an image sensor device of the computing device, a plurality of responsive eye states each corresponding to respective ones of the plurality of temporal instances of the first time period;

wherein each of the plurality of responsive eye states includes one or more pairs, each pair comprising a user eye location and a corresponding RGB value measured at the user eye location,

wherein each of the corresponding RGB values are reflections caused by interactions between the display pattern with a user eye at the user eye location,

determine a responsive correlation scorer by:

for the each of the plurality of temporal instances,

retrieving a corresponding displayed image from the series of displayed images,

retrieving an expected pair comprising an expected user eye location and an expected RGB value based at least on the corresponding displayed image, the user, and at least one of a database or a predictive model,

using the expected user eye location, identify at least one of the one or more pairs having an associated user eye location correlating to the expected user eye location, and

comparing an associated RGB value of the identified at least one of the one or more pairs to the expected RGB value,

and determine whether to grant the user authentication request based on the responsive correlation score.

2. The apparatus of claim 1 , wherein generating the display pattern comprises:

identifying a plurality of pixels characterizing a display space of the display device;

for each temporal instance of the plurality of temporal instances in the first time period,

selecting, from the plurality of pixels, one or more selected pixels for a temporal instance using a first pseudo-random selection;

selecting, for each selected pixel of the one or more selected pixels for the temporal instance, a corresponding color designation of a plurality of color designations using a second pseudo-random selection; and

associating each selected pixel of the one or more selected pixels for the temporal instance with the corresponding color designation for the selected pixel to generate a per-temporal-instance display pattern for the temporal instance; and

generating the display pattern based on each per-temporal-instance display pattern for each temporal instance of the plurality of temporal instances in the first time period.

3. The apparatus of claim 1 , wherein:

causing the display device of the computing device to display the display pattern at the first time period comprises causing the display device to display the display pattern at a first subperiod within the first time period; and

the first time period comprises the first subperiod and a responsive time period after a first temporal instance.

4. The apparatus of claim 3 , wherein the responsive time period is selected based on a responsive behavioral profile of the user.

5. The apparatus of claim 3 , wherein the responsive time period is selected based on a responsive behavioral profile of an average end user.

6. The apparatus of claim 1 , wherein determining the responsive eye state of the user comprises:

detecting, based on the image data, one or more responsive eye-related facial features of the user; and

determining the responsive eye state based on the one or more responsive eye-related facial features.

7. The apparatus of claim 1 , wherein:

the responsive correlation score further comprises a lip synchronization correlation score; and

determining the lip synchronization correlation score comprises:

determining, based on second image sensor data received from an image sensor device of the computing device, a lip synchronization state of the user at a second time period; and

determining the lip synchronization correlation score based on a second estimated likelihood that the lip synchronization state corresponds to an expected lip synchronization state, wherein the expected lip synchronization state is determined using a generative machine learning model.

8. The apparatus of claim 7 , wherein the expected lip synchronization state comprises expected lip movements while uttering an authentication token.

9. The apparatus of claim 8 , wherein the authentication token is a preconfigured password of a user account associated with the user authentication request.

10. The apparatus of claim 1 , wherein determining the responsive correlation score for the display pattern and the responsive eye state comprises:

for each temporal instance of the plurality of temporal instances of the first time period,

determining a per-temporal-instance display pattern of the display pattern at a temporal instance;

identifying a reactive temporal instance of the plurality of temporal instances for the temporal instance, wherein the reactive temporal instance for the temporal instance has a temporal distance from the temporal instance that corresponds to a responsive time period;

determining a reactive subset of the responsive eye state for the temporal instance, wherein the reactive subset comprises a portion of the responsive eye state determined based on the image sensor data whose timestamp corresponds to the reactive temporal instance; and

determining a per-temporal-instance responsive correlation score for the temporal instance based on a measure of deviation of the per-temporal-instance display pattern from the reactive subset; and

determining the responsive correlation score based on each per-temporal-instance responsive correlation score for each temporal instance of the plurality of temporal instances.

11. The apparatus of claim 1 , whether the display pattern is a pseudo-random generated display pattern.

12. A computer-implemented method for illumination-based user authentication, the computer-implemented method comprising:

receiving a user authentication request from a user for a computing device,

and in response to receiving the user authentication request,

causing a display device of the computing dev ice to display a display pattern comprising a series of displayed images during a first time period comprising a plurality of temporal instances,

wherein for each of the plurality of temporal in stances, a different one of the series of displayed images is displayed on the display device,

wherein each of the different one of the series of displayed images corresponds to a defined combination of Red-Green-Blue (RGB) color and position on the display,

determining, based on image sensor data received from an image sensor device of the computing device, a plurality of responsive eye states, each corresponding to respective ones of the plurality of temporal instances of the first time period;

wherein each of the plurality of responsive eye states includes one or more pairs,

each pair comprising a user eye location and a corresponding RGB value measured at the user eye location,

wherein each of the corresponding RGB values are reflections caused by interactions between the display pattern with a user eye at the user eye location,

determining a responsive correlation score by:

for the each of the plurality of temporal instances,

retrieving a corresponding displayed image from the series of displayed images,

retrieving an expected pair comprising an expected user eye location and

an expected RGB value based at least on the corresponding displayed image, the user, and at least one of a database or a predictive model,

using the expected user eye location, identify at least one of the one or more pairs having an associated user eye location correlating to the expected user eye location, and

comparing an associated RGB value of the identified at least one of the one or more pairs to the expected RGB value,

and determining whether to grant the user authentication request based on the responsive correlation score.

13. The computer-implemented method of claim 12 , wherein generating the display pattern comprises:

identifying a plurality of pixels characterizing a display space of the display device;

for each temporal instance of the plurality of temporal instances in the first time period,

selecting, from the plurality of pixels, one or more selected pixels for a temporal instance using a first pseudo-random selection;

selecting, for each selected pixel of the one or more selected pixels for the temporal instance, a corresponding color designation of a plurality of color designations using a second pseudo-random selection; and

associating each selected pixel of the one or more selected pixels for the temporal instance with the corresponding color designation for the selected pixel to generate a per-temporal-instance display pattern for the temporal instance; and

generating the display pattern based on each per-temporal-instance display pattern for each temporal instance of the plurality of temporal instances in the first time period.

14. The computer-implemented method of claim 12 , wherein:

causing the display device of the computing device to display the display pattern at the first time period comprises causing the display device to display the display pattern at a first subperiod within the first time period; and

the first time period comprises the first subperiod and a responsive time period after a first temporal instance.

15. The computer-implemented method of claim 12 , wherein determining the responsive eye state of the user comprises:

detecting, based on the image data, one or more responsive eye-related facial features of the user; and

determining the responsive eye state based on the one or more responsive eye-related facial features.

16. The computer-implemented method of claim 12 , wherein:

the responsive correlation score further comprises a lip synchronization correlation score; and

determining the lip synchronization correlation score comprises:

determining, based on second image sensor data received from an image sensor device of the computing device, a lip synchronization state of the user at a second time period; and

determining the lip synchronization correlation score based on a second estimated likelihood that the lip synchronization state corresponds to an expected lip synchronization state, wherein the expected lip synchronization state is determined using a generative machine learning model.

17. The computer-implemented method of claim 12 , wherein determining the responsive correlation score for the display pattern and the responsive eye state comprises:

for each temporal instance of a plurality of temporal instances of the first time period,

determining a per-temporal-instance display pattern of the display pattern at a temporal instance;

identifying a reactive temporal instance of the plurality of temporal instances for the temporal instance, wherein the reactive temporal instance for the temporal instance has a temporal distance from the temporal instance that corresponds to a responsive time period;

determining a reactive subset of the responsive eye state for the temporal instance, wherein the reactive subset comprises a portion of the responsive eye state determined based on the image sensor data whose timestamp corresponds to the reactive temporal instance; and

determining a per-temporal-instance responsive correlation score for the temporal instance based on a measure of deviation of the per-temporal-instance display pattern from the reactive subset; and

determining the responsive correlation score based on each per-temporal-instance responsive correlation score for each temporal instance of the plurality of temporal instances.

18. The computer-implemented method of claim 12 , wherein the display pattern is a pseudo-random-generated display pattern.

19. A computer program product for illumination-based user authentication the computer program product comprising at least one non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions configured to:

receive a user authentication request from a user for a computing device,

and in response to receiving the user authentication request,

cause a display device of the computing device to display a display pattern comprising a series of displayed images during a first time period comprising a plurality of temporal instances,

wherein for each of the plurality of temporal in stances, a different one of the series of displayed images is displayed on the display device,

wherein each of the different one of the series of displayed images corresponds to a defined combination of Red-Green-Blue (RGB) color and position on the display,

determine, based on image sensor data received from an image sensor device of the computing device, a plurality of responsive eye states, each corresponding to respective ones of the plurality of temporal instances of the first time period;

wherein each of the plurality of responsive eye states includes one or more pairs, each pair comprising a user eye location and a corresponding RGB value measured at the user eye location,

wherein each of the corresponding RGB values are reflections caused by interactions between the display pattern with a user eye at the user eye location,

determine a responsive correlation score by:

for the each of the plurality of temporal instances,

retrieving a corresponding displayed image from the series of displayed images,

retrieving an expected pair comprising an expected user eye location and an expected RGB value based at least on the corresponding displayed image, the user, and at least one of a database or a predictive model,

using the expected user eye location, identify at least one of the one or more pairs having an associated user eye location correlating to the expected user eye location, and

comparing an associated RGB value of the identified at least one of the one or more pairs to the expected RGB value,

and determine whether to grant the user authentication request based on the responsive correlation score.

20. The computer program product of claim 19 , wherein determining the responsive correlation score for the display pattern and the responsive eye state comprises:

for each temporal instance of the plurality of temporal instances of the first time period,

determining a per-temporal-instance display pattern of the display pattern at a temporal instance;

identifying a reactive temporal instance of the plurality of temporal instances for the temporal instance, wherein the reactive temporal instance for the temporal instance has a temporal distance from the temporal instance that corresponds to a responsive time period;

determining a reactive subset of the responsive eye state for the temporal instance, wherein the reactive subset comprises a portion of the responsive eye state determined based on the image sensor data whose timestamp corresponds to the reactive temporal instance; and

determining a per-temporal-instance responsive correlation score for the temporal instance based on a measure of deviation of the per-temporal-instance display pattern from the reactive subset; and

determining the responsive correlation score based on each per-temporal-instance responsive correlation score for each temporal instance of the plurality of temporal instances.