State estimation device, method and computer program therefor
A state of a user of a mobile terminal having a display for displaying an image may be estimated. A sightline detector detects a sightline of the user of a mobile terminal and processing circuitry is configured to estimate a state of the user. The processing circuitry is configured to determine a top-down index value that is correlated with an attention source amount allocated to top-down attention of the user with respect to an image displayed on a display of the mobile terminal. The processing circuitry is configured to determine a bottom-up index value correlated with the attention source amount allocated to bottom-up attention of the user with respect to the image displayed on the display. The processing circuitry is configured to estimate the user state including an attention function degraded state based on the top-down index value and the bottom-up index value.
1. A device that estimates a state of a user of a mobile terminal having a display for displaying an image, the state estimation device comprising
a sightline detector that detects a sightline of the user; and
processing circuitry configured to:
determine a top-down index value based on the image displayed on the display and the sightline of the user detected by the sightline detector, the top-down index value being correlated with an attention source amount that is allocated to top-down attention of the user with respect to the image on the display;
determine a bottom-up index value based on the image displayed on the display and the sightline of the user detected by the sightline detection, the bottom-up index value being correlated with the attention source amount that is allocated to bottom-up attention of the user with respect to the image on the display; and
estimate the state of the user including an attention function degraded state based on the top-down index value and the bottom-up index value, wherein
on condition that the top-down index value falls below a predetermined top-down threshold, the processing circuitry is configured to:
estimate that the user is in a first attention function degraded state when the bottom-up index value does not fall below a predetermined bottom-up threshold, and
estimate that the user is in a second attention function degraded state, in which an attention function is degraded in comparison with that in the first attention function degraded state, when the bottom-up index value falls below the bottom-up threshold.
2. The device according to claim 1 , wherein
the processing circuitry is configured to
determine a point of interest in the image displayed on the display; and
further determine the top-down index value based on motion of the sightline of the user with respect to the point of interest in the image.
3. The device according to claim 2 , wherein
the processing circuitry configured to:
determine distribution of saliency in the image displayed on the display; and
further determine the bottom-up index value based on motion of the sightline of the user with respect to the distribution of the saliency in the image.
4. The device according to claim 1 , wherein
the processing circuitry is configured to further estimate the user state based on one of the image displayed on the display or a surrounding environment, and the sightline of the user detected by the sightline detector and a cognitive response time of the user.
5. The device according to claim 4 , wherein
the processing circuitry is configured to selectively further estimate the user state based on one of the image displayed on the display or a surrounding environment according to likeliness of occurrence of bottom-up attention of the user with respect to the image displayed on the display.
6. The device according to claim 1 , wherein the attention function degraded state includes:
a first attention function degraded state that is at least one of wakefulness degradation, fatigue, and a drifted state; and
a second attention function degraded state that is at least one of a cardiac disorder, a brain disorder, epilepsy, and hypoglycemia.
7. The device according to claim 3 , wherein
the processing circuitry is configured to further estimate the user state based on one of the image displayed on the display or a surrounding environment, and the sightline of the user detected by the sightline detector and a cognitive response time of the user.
8. The device according to claim 7 , wherein
the processing circuitry is configured to selectively further estimate the user state based on one of the image displayed on the display or a surrounding environment according to likeliness of occurrence of bottom-up attention of the user with respect to the image displayed on the display.
9. The device according to claim 2 , wherein
the processing circuitry is configured to further estimate the user state based on one of the image displayed on the display or a surrounding environment, and the sightline of the user detected by the sightline detector and a cognitive response time of the user.
10. The device according to claim 1 , wherein
the processing circuitry configured to:
determine distribution of saliency in the image displayed on the display; and
further determine the bottom-up index value based on motion of the user's sightline with respect to the distribution of the saliency in the image.
11. The device according to claim 10 , wherein
the processing circuitry is configured to further estimate the user state based on one of the image displayed on the display or a surrounding environment, and the sightline of the user detected by the sightline detector and a cognitive response time of the user.
12. The device according to claim 1 , wherein the processing circuitry is configured to selectively further estimate the user state based on one of the image displayed on the display or a surrounding environment according to likeliness of occurrence of bottom-up attention of the user with respect to the image displayed on the display.
13. A method of estimating a state of a user of a mobile terminal having a display for displaying an image, the method comprising:
detecting a sightline of the user;
determining a top-down index value based on the image displayed on the display and the sightline of the user, the top-down index value being correlated with an attention source amount that is allocated to top-down attention of the user with respect to the image on the display;
determining a bottom-up index value based on the image displayed on the display and the sightline of the user detected by the sightline detection, the bottom-up index value being correlated with the attention source amount that is allocated to bottom-up attention of the user with respect to the image on the display; and
estimating the state of the user including an attention function degraded state based on the top-down index value and the bottom-up index value, wherein
on condition that the top-down index value falls below a predetermined top-down threshold,
estimating that the user is in a first attention function degraded state when the bottom-up index value does not fall below a predetermined bottom-up threshold, and
estimating that the user is in a second attention function degraded state, in which an attention function is degraded in comparison with that in the first attention function degraded state, when the bottom-up index value falls below the bottom-up threshold.
14. A non-transitory computer readable storage device having computer readable instructions that when executed by a controller including a computer cause the computer to execute a state of a user of a mobile terminal estimation method, the method comprising:
detecting a sightline of the user;
determining a top-down index value based on an image displayed on the display and the sightline of the user detected, the top-down index value being correlated with an attention source amount that is allocated to top-down attention of the user with respect to the image on the display;
determining a bottom-up index value based on the image displayed on the display and the sightline of the user detected by the sightline detection, the bottom-up index value being correlated with the attention source amount that is allocated to bottom-up attention of the user with respect to the image on the display; and
estimating the state of the user including an attention inaction degraded state based on the top-down index value and the bottom-up index value, wherein
on condition that the top-down index value falls below a predetermined top-down threshold,
estimating that the user is in a first attention function degraded state when the bottom-up index value does not fall below a predetermined bottom-up threshold, and
estimating that the user is in a second attention function degraded state, in which an attention function is degraded in comparison with that in the first attention function degraded state, when the bottom-up index value falls below the bottom-up threshold.