Identity authentication method, identity authentication device, and terminal

1. An identity authentication method, comprising:

providing an authentication object with prompt information, wherein the prompt information is used to prompt the authentication object to make a target authentication action, wherein the target authentication action comprises at least one of a facial expression and a head action;

collecting at least one frame of image;

detecting whether authentication action information of the authentication object in the at least one frame of image matches reference action information, wherein the reference action information is information generated when a target object makes the target authentication action;

determining that the authentication object passes an identity authentication in a case that the authentication action information of the authentication object matches the reference action information;

acquiring a three-dimensional head model of the target object; and

obtaining the reference action information generated when the target object makes the target authentication action by simulating based on the three-dimensional head model and a preset moving trajectory corresponding to a head during the head moving, wherein the reference action information is represented by at least one of an image and a feature point.

2. The identity authentication method according to claim 1 , wherein the providing the authentication object with the prompt information comprises:

providing the prompt information to the authentication object randomly, wherein the prompt information corresponds to at least one authentication action.

3. The identity authentication method according to claim 1 , wherein the providing the authentication object with the prompt information comprises:

acquiring the target authentication action randomly from a plurality of authentication actions;

generating the prompt information based on the target authentication action; and

providing the prompt information to the authentication object.

4. The identity authentication method according to claim 1 , wherein the acquiring the three-dimensional head model of the target object comprises:

collecting images comprising the head of the target object at different angles; and

obtaining the three-dimensional head model by training based on the collected images.

5. The identity authentication method according to claim 1 , wherein the obtaining the reference action information generated when the target object makes the target authentication action by simulating based on the three-dimensional head model and the preset moving trajectory corresponding to the head during the head moving comprises:

determining feature points of the head of the target object based on the three-dimensional head model; and

obtaining the reference action information by simulating based on the feature points and the preset moving trajectory.

6. The identity authentication method according to claim 1 , further comprising:

collecting an image of a head of the target object making the target authentication action; and

storing action information in the collected image as the reference action information.

7. An identity authentication device, comprising:

an information providing module, configured to provide an authentication object with prompt information, wherein the prompt information is used to prompt the authentication object to make a target authentication action, wherein the target authentication action comprises at least one of a facial expression and a head action;

a first collecting module, configured to collect at least one frame of image;

a detecting module, configured to detect whether authentication action information of the authentication object in the at least one frame of image matches reference action information, wherein the reference action information is information generated when a target object makes the target authentication action;

a result determining module, configured to determine that the authentication object passes an identity authentication in a case that a detection result of the detecting module indicates that the authentication action information of the authentication object matches the reference action information;

a model acquiring module, configured to acquire a three-dimensional head model of the target object; and

an information simulating module, configured to obtain the reference action information generated when the target object makes the target authentication action by simulating based on the three-dimensional head model and a preset moving trajectory corresponding to a head during the head moving, wherein the reference action information is represented by at least one of an image and a feature point.

8. The identity authentication device according to claim 7 , wherein the information providing module is further configured to:

provide the prompt information to the authentication object randomly, wherein the prompt information corresponds to at least one authentication action.

9. The identity authentication device according to claim 7 , wherein the information providing module comprises:

an acquiring unit, configured to acquire the target authentication action randomly from a plurality of authentication actions;

a generating unit, configured to generate the prompt information based on the target authentication action; and

a providing unit, configured to provide the prompt information to the authentication object.

10. The identity authentication device according to claim 7 , wherein the model acquiring module comprises:

a collecting unit, configured to collect images comprising the head of the target object at different angles; and

a training unit, configured to obtain the three-dimensional head model by training based on the collected images.

11. The identity authentication device according to claim 7 , wherein the information simulating module comprises:

a determining unit, configured to determine feature points of the head of the target object based on the three-dimensional head model; and

a simulating unit, configured to obtain the reference action information by simulating based on the feature points and the preset moving trajectory.

12. The identity authentication device according to claim 7 , further comprising:

a second collecting module, configured to collect an image of a head of the target object making the target authentication action; and

an information storing module, configured to store action information in the collected image as the reference action information.

13. A terminal, comprising:

one or more processors; and

a memory stored with one or more programs configured to be executed by the one or more processors, wherein the one or more programs comprise an instruction to:

provide an authentication object with prompt information, wherein the prompt information is used to prompt the authentication object to make a target authentication action, wherein the target authentication action comprises at least one of a facial expression and a head action;

collect at least one frame of image;

detect whether authentication action information of the authentication object in the at least one frame of image matches reference action information, wherein the reference action information is information generated when a target object makes the target authentication action;

determine that the authentication object passes an identity authentication in a case that the authentication action information of the authentication object matches the reference action information;

acquire a three-dimensional head model of the target object; and

obtain the reference action information generated when the target object makes the target authentication action by simulating based on the three-dimensional head model and a preset moving trajectory corresponding to a head during the head moving, wherein the reference action information is represented by at least one of an image and a feature point.

14. The terminal according to claim 13 , wherein the one or more programs further comprise an instruction to:

provide the prompt information to the authentication object randomly, wherein the prompt information corresponds to at least one authentication action.

15. The terminal according to claim 13 , wherein the one or more programs further comprise an instruction to:

acquire the target authentication action randomly from a plurality of authentication actions;

generate the prompt information based on the target authentication action; and

provide the prompt information to the authentication object.

16. The terminal according to claim 13 , wherein the one or more programs further comprise an instruction to:

collect images comprising the head of the target object at different angles; and

obtain the three-dimensional head model by training based on the collected images.

17. The terminal according to claim 13 , wherein the one or more programs further comprise an instruction to:

determine feature points of the head of the target object based on the three-dimensional head model; and

obtain the reference action information by simulating based on the feature points and the preset moving trajectory.

18. The terminal according to claim 13 , wherein the one or more programs further comprise an instruction to:

collect an image of a head of the target object making the target authentication action; and

store action information in the collected image as the reference action information.