Method and system for monitoring an occupant of a vehicle

1. A method comprising:

detecting an occupant in at least one image, the detection sufficient to associate a stored user profile with the occupant;

identifying body points of the occupant based on the at least one image;

classifying a pose of the occupant based on the identified body points;

determining, via a neural network, a postural comfort of the occupant based on the classified pose of the occupant and at least one of a movement of the occupant or a facial expression of the occupant, the neural network trained to determine the postural comfort of the occupant by estimating a seating pose comfort score;

determining a remaining trip duration based on navigation information;

generating, via a recommendation engine, a recommendation for adapting the pose of the occupant based on the postural comfort of the occupant and the remaining trip duration; and

providing the recommendation for adapting the pose of the occupant based on at least one of preferences in the stored user profile or a road condition, the recommendation provided as a visual response including at least one of:

an icon shown on a head-up display of a vehicle;

an icon shown on a screen of an infotainment system of the vehicle;

a dialog shown on the head-up display of the vehicle; or

a dialog shown on the screen of the infotainment system of the vehicle.

2. The method according to claim 1 ,

wherein the at least one image comprises an image sequence, and wherein the calculation of the seating pose comfort score is based on the classified pose and the movement of the occupant, the method further comprising:

determining, responsive to identifying body points, a temporal sequence of the body points based on the image sequence;

estimating a displacement vectors for each of the body points based on the temporal sequence; and

classifying the movement of the occupant based on the displacement vectors; and

wherein the recommendation for adapting the pose of the occupant is provided further based on the classified movement of the occupant.

3. The method according to claim 2 ,

wherein the body points include facial points which are identified and tracked based on the image sequence, and wherein the calculation of the seating pose comfort score is based on the facial expression of the occupant, the method further comprising:

classifying, responsive to identifying body points, the facial expression based on the identified facial points; and

wherein the recommendation for adapting the pose of the occupant is provided further based on the classified facial expression.

4. The method according to claim 1 , wherein the recommendation for adapting the pose of the occupant is provided only if the seating pose comfort score is less than a predetermined threshold.

5. The method according to claim 1 , further comprising:

monitoring a position of the occupant over a predefined time period based on the identified body points at a beginning and at an end of the predefined time period; and

providing the recommendation for adapting the occupant's pose if the occupant's position has not changed during the predefined time period.

6. The method according to claim 1 ,

wherein the occupant is located on a driver seat; and

wherein the body points include at least one of shoulder points, arm points, or hand points.

7. The method according to claim 6 , further comprising:

determining a relative position of the at least one shoulder points, arm points, or hand points with respect to a steering wheel of the vehicle; and

providing, based on the relative position, a further recommendation for adapting the steering wheel.

8. The method according to claim 1 , further comprising:

acquiring a response of the occupant to the recommendation, the response including at least one of accepting or rejecting the recommendation; and

determining one or more of a quantity and a type of future recommendations in accordance with the response of the occupant.

9. The method according to claim 8 , wherein if the occupant accepts the recommendation, the method further comprises at least one of:

proposing guided activities for the occupant which depend on a driving condition and an environment of the vehicle; or

activating a massage seat function.

10. The method according to claim 1 , further comprising:

classifying the pose of the occupant further based on a stored user profile of associated with the occupant.

11. The method according to claim 1 , wherein the stored user profile comprises at least one recorded seating pose comfort score previously calculated.

12. A system comprising:

a sensing device configured to detect an occupant in at least one image;

a processing unit configured to:

associate a stored user profile with the occupant based on the detection of the occupant;

identify body points of the occupant based on the at least one image;

classify a pose of the occupant based on the identified body points;

determine, via a neural network, a postural comfort of the occupant based on the classified pose of the occupant and at least one of a movement of the occupant or a facial expression of the occupant, the neural network trained to determine the postural comfort of the occupant by estimating a seating pose comfort score;

determine a remaining trip duration based on navigation information; and

generate, via a recommendation engine, a recommendation for adapting the pose of the occupant based on the postural comfort of the occupant and the remaining trip duration; and

an output unit configured to provide the recommendation for adapting the pose of the occupant based on at least one of preferences in the stored user profile or a road condition, the recommendation provided as a visual response including at least one of:

an icon shown on a head-up display of a vehicle;

an icon shown on a screen of an infotainment system of the vehicle;

a dialog shown on the head-up display of the vehicle; or

a dialog shown on the screen of the infotainment system of the vehicle.

13. The system according to claim 12 , wherein the system further comprises:

an input unit configured to acquire a response of the occupant to the recommendation, the response including at least one of accepting or rejecting the recommendation and one or more of a quantity and a type of future recommendations depends on the response.

14. The system according to claim 12 , wherein the processing unit includes at least one of a machine-learning algorithm or a rule-based logic for classifying the pose of the occupant.

15. The system according to claim 12 ,

wherein the sensing device is configured to detect an image sequence,

wherein the seating pose comfort score is based on the classified pose of the occupant and the movement of the occupant, and

wherein the processing unit is configured to:

determine, responsive to identification of body points, a temporal sequence of the body points based on the image sequence,

estimate displacement vectors for each of the body points using the temporal sequence, and

classify the movement of the occupant based on the displacement vectors, and

wherein the output unit provides the recommendation for adapting the pose of the occupant is further based on the classified movement of the occupant.

16. The system according to claim 15 ,

wherein the body points include facial points which are identified and tracked based on the image sequence,

wherein the calculation of the seating pose comfort score is further based on the facial expression of the occupant, and

wherein the processing unit is configured to:

track the body points based on the image sequence,

derive a facial expression based on the identified facial points, and

classify facial expressions based on the identified facial points, and

wherein the output unit is configured to provide the recommendation for adapting the pose of the occupant is further based on one or more classified facial expressions.

17. The system according to claim 15 , wherein the output unit is configured to provide the recommendation for adapting the pose of the occupant only if the seating pose comfort score is less than a predetermined threshold.

18. The system according to claim 12 ,

wherein the processing unit is configured to monitor a position of the occupant over a predefined time period by identifying the body points at a beginning and at an end of the predefined time period, and

wherein the output unit is configured to provide the recommendation for adapting the pose of the occupant if the position of the occupant has not changed during the predefined time period.

19. The system according to claim 12 ,

wherein the occupant is located on a driver seat,

wherein the body points include at least one of shoulder points, arm points, or hand points,

wherein the processing unit is configured to determine a relative position of the at least one shoulder points, arm points, or hand points with respect to a steering wheel of the vehicle, and

wherein the output unit is configured to provide, based on the relative position, a further recommendation for adapting the steering wheel.

20. The system according to claim 12 , wherein the stored user profile comprises at least one recorded seating pose comfort score previously calculated.