Method and apparatus for eveluating health condition by using skeleton model

1 . A method for evaluating health status using skeleton models, performed by a health management server, the method comprising the steps of:

providing a user device with a standard motion image of a trainer making a standard motion;

receiving a user motion image for a user motion of a user imitating the standard motion from the user device;

calculating a similarity between the standard motion and the user motion by comparing the user motion image with the standard motion image; and

evaluating the health status of the user based on the calculated similarity,

wherein the step of calculating the similarity comprises the steps of:

acquiring a trainer skeleton motion model for the trainer making the standard motion in the standard motion image;

acquiring a user skeleton motion model for the user making the user motion in the user motion image; and

calculating the similarity by comparing feature points for each corresponding body part between the trainer skeleton motion model and the user skeleton motion model, wherein the feature points for each body part are a plurality of nodes of the skeleton motion models,

wherein the step of calculating the similarity by comparing the feature points for each body part comprises the steps of:

fetching a weight per body part for the standard motion from a database; and

applying the weight per body part to the similarity calculation,

wherein the step of evaluating the health status comprises evaluating exhaustion of physical strength and cognitive impairment level of the user,

wherein

the standard motion includes a plurality of standard motions, and the cognitive impairment type includes a plurality of cognitive impairment types,

the cognitive impairment level is evaluated based on behavior feature data in cognitive impairment type for each standard motion of the plurality of standard motions,

the weight per body part is assigned for each standard motion of the plurality of standard motions, and the weight per body part is subdivided for each cognitive impairment type of the plurality of cognitive impairment types, and

the behavior feature data is data of corresponding nodes of the plurality of nodes representative of behavioral features of the user for each cognitive impairment type for each standard motion.

2 . The method for evaluating health status using skeleton models according to claim 1 , wherein the exhaustion of physical strength is determined based on a sum of movement trajectories of the feature points for each body part of the user skeleton motion model; and a predefined physical strength exhaustion evaluation weight for an amount of movement for each body part.

3 . The method for evaluating health status using skeleton models according to claim 1 , wherein the method comprises the steps of:

receiving the user motion image for each of a plurality of users from a plurality of user devices;

displaying the plurality of received user motion images on a display device;

calculating the similarity between each of the plurality of user motions in the plurality of received user motion images and the standard motion; and

highlighting at least one of the plurality of user motion images displayed on the display device based on the calculated similarity.

4 . The method for evaluating health status using skeleton models according to claim 3 , wherein the step of highlighting comprises highlighting a predetermined number of user motion images having the calculated similarity in a lower or higher range than a predetermined criterion among the plurality of user motion images.

5 . The method for evaluating health status using skeleton models according to claim 3 , wherein the step of highlighting comprises the steps of:

calculating the similarity between users by comparing the skeleton motion models for the plurality of user motions;

grouping the plurality of user motion images based on the calculated similarity between users; and

differently displaying the grouped user motion images.

6 . The method for evaluating health status using skeleton models according to claim 5 , wherein the step of calculating the similarity between users by comparing the skeleton motion models for the plurality of user motions comprises not applying the weight per body part for the standard motion.

7 . A non-transitory computer readable recording medium including executable instructions, wherein the instructions, when executed by a processor, cause the processor to perform the method according to claim 1 .