Detecting facial liveliness

1. A computer-implemented method of detecting facial liveliness, comprising:

processing, by one or more processors, first and second facial images of a subject to determine first and second poses of a face of the subject, the first and second facial images being captured at first and second sequential time points by a camera device;

determining, by the one or more processors, a motion based on information of a relative movement between the camera device and the face of the subject, wherein the relative movement is between the first and second time points;

determining, by the one or more processors, a change in pose of the face based on the determined first and second poses;

comparing, by the one or more processors, the determined change in pose of the face to the motion; and

determining, by the one or more processors, facial liveliness of the subject based on a result of the comparison.

2. The method of claim 1 , wherein comparing the determined change in pose of the face to a motion comprises:

correlating the determined change in pose of the face to the motion;

scoring a matching quality based on a result of the correlation; and

comparing the scored matching quality to a predetermined threshold, and

wherein determining facial liveliness of the subject comprises:

determining that the face of the subject is live in response to determining that the scored matching quality is beyond the predetermined threshold.

3. The method of claim 2 , wherein the predetermined threshold is based on a varying context-dependent requirement.

4. The method of claim 1 , further comprising: prompting the subject to move the camera device relative to the face of the subject.

5. The method of claim 4 , further comprising:

receiving information of the movement of the camera device measured by a sensor; and

determining the motion based on the received information of the movement of the camera device.

6. The method of claim 5 , wherein the one or more processors are included in a computing device, and

wherein the computing device includes the camera device and the sensor, and the sensor includes at least one of an accelerometer, a gyroscope, or a global positioning system (GPS).

7. The method of claim 1 , further comprising: determining the motion by analyzing corneal reflections of the camera device in the first and second facial images.

8. The method of claim 1 , wherein determining first and second poses of the face comprises analyzing one or more facial landmarks of the face.

9. The method of claim 8 , wherein analyzing one or more facial landmarks of the face comprises determining variance of orientation and size of the one or more facial landmarks.

10. The method of claim 9 , wherein each of the facial landmarks is associated with a respective commensurate change in orientation and size, and

wherein the motion is based on a combination of expected respective commensurate changes of the facial landmarks.

11. The method of claim 8 , wherein the one or more facial landmarks comprise one or more of nose, forehead, eye brows, eyes, eye corners, lips, and mouths.

12. The method of claim 1 , further comprising:

processing the first and second facial images of the subject to determine first and second corneal reflections of an object;

determining a corneal reflection change of the object based on the determined first and second corneal reflections;

comparing the determined corneal reflection change of the object to an expected corneal reflection change based on the motion; and

determining the facial liveliness of the subject based on both of a result of the comparison of the determined corneal reflection change to the expected corneal reflection change and the result of the comparison of the determined change in pose of the face to the motion.

13. A computer-implemented method of detecting facial liveliness, comprising:

processing, by one or more processors, first and second facial images of a subject to determine first and second poses of a face of the subject, the first and second facial images being captured at first and second sequential time points by a camera device;

determining, by the one or more processors, a motion based on information of a relative movement between the camera device and the face of the subject, wherein the relative movement is between the first and second sequential time points;

calculating, by the one or more processors, an expected pose of the face at the second sequential time point based on the determined first pose and the motion; and

assessing, by the one or more processors, facial liveliness of the subject by determining likelihood between the calculated expected pose of the face to the determined second pose of the face from the second facial image.

14. The method of claim 13 , wherein assessing the facial liveliness of the subject comprises:

comparing a correlation between the determined second pose of the face and the calculated expected pose of the face to a predetermined threshold; and

determining whether the face of the subject is live based on a result of the comparison,

wherein assessing the facial liveliness of the subject comprises:

determining that the face of the subject is live in response to determining that the correlation is beyond the determined threshold.

15. The method of claim 14 , wherein the predetermined threshold is based on a varying context-dependent requirement.

16. A computer-implemented method for detecting facial liveliness, comprising:

transmitting, by one or more processors, a command to a camera device to capture facial images of a subject at sequential time points including first and second sequential time points;

receiving, by the one or more processors, the captured facial images from the camera device;

processing, by one or more processors, first and second facial images of the subject in the captured facial images to determine first and second poses of a face of the subject, the first and second facial images being captured at first and second sequential time points;

determining, by the one or more processors, a change in pose of the face based on the determined first and second poses;

determining, by the one or more processors, a motion based on information of a relative movement between the camera device and the face of the subject, wherein the relative movement is between the first and second sequential time points;

comparing, by the one or more processors, the determined change in pose of the face to the motion; and

determining, by the one or more processors, facial liveliness of the subject based on a result of the comparison.

17. The method of claim 16 , wherein the camera device is moved from a first position to a second position between the first and second time points,

wherein the first facial image of the subject is captured at the first time point when the camera device is at the first position, and the second facial image of the subject is captured at the second time point when the camera device is at the second position, and

wherein the motion is based on a movement of the camera device between the first and second positions.

18. The method of claim 17 , further comprising:

transmitting a second command to a controller coupled to an object, the second command indicating the controller to move the object at the first and second time points,

wherein the motion is based on the movement of the camera device and the movement of the object.

19. A computer-implemented method of detecting facial liveliness, comprising:

processing, by one or more processors, first and second facial images of a subject to determine first and second poses of a face of the subject, the first and second facial images being captured at first and second sequential time points;

determining, by the one or more processors, a change in pose of the face based on the determined first and second poses;

comparing, by the one or more processors, the determined change in pose of the face to a motion associated with the first and second time points; and

determining, by the one or more processors, facial liveliness of the subject based on a result of the comparison,

wherein comparing the determined change in pose of the face to a motion comprises:

correlating the determined change in pose of the face to the motion;

scoring a matching quality based on a result of the correlation; and

comparing the scored matching quality to a predetermined threshold, and

wherein determining facial liveliness of the subject comprises:

determining that the face of the subject is live in response to determining that the scored matching quality is beyond the predetermined threshold.

20. A computer-implemented method of detecting facial liveliness, comprising:

processing, by one or more processors, first and second facial images of a subject to determine first and second poses of a face of the subject, the first and second facial images being captured at first and second sequential time points;

determining, by the one or more processors, a change in pose of the face based on the determined first and second poses;

determining a motion by analyzing corneal reflections of a camera device in the first and second facial images, wherein the motion is associated with a relative movement between the face and the camera device configured to capture facial images of the subject;

comparing, by the one or more processors, the determined change in pose of the face to the motion associated with the first and second time points; and

determining, by the one or more processors, facial liveliness of the subject based on a result of the comparison.

21. A computer-implemented method of detecting facial liveliness, comprising:

processing, by one or more processors, first and second facial images of a subject to determine first and second poses of a face of the subject, the first and second facial images being captured at first and second sequential time points;

determining, by the one or more processors, a change in pose of the face based on the determined first and second poses;

comparing, by the one or more processors, the determined change in pose of the face to a motion associated with the first and second time points; and

determining, by the one or more processors, facial liveliness of the subject based on a result of the comparison,

wherein determining first and second poses of the face comprises analyzing one or more facial landmarks of the face, and wherein analyzing one or more facial landmarks of the face comprises determining variance of orientation and size of the one or more facial landmarks.

22. A computer-implemented method of detecting facial liveliness, comprising:

processing, by one or more processors, first and second facial images of a subject to determine first and second poses of a face of the subject, the first and second facial images being captured at first and second sequential time points;

determining, by the one or more processors, a change in pose of the face based on the determined first and second poses;

comparing, by the one or more processors, the determined change in pose of the face to a motion associated with the first and second time points;

processing the first and second facial images of the subject to determine first and second corneal reflections of an object;

determining a corneal reflection change of the object based on the determined first and second corneal reflections;

comparing the determined corneal reflection change of the object to an expected corneal reflection change based on the motion; and

determining facial liveliness of the subject based on both of a result of the comparison of the determined change in pose of the face to the motion and a result of the comparison of the determined corneal reflection change to the expected corneal reflection change.

23. A computer-implemented method of detecting facial liveliness, comprising:

processing, by one or more processors, first and second facial images of a subject to determine first and second poses of a face of the subject, the first and second facial images being captured at first and second sequential time points;

calculating, by the one or more processors, an expected pose of the face at the second sequential time point based on the determined first pose and a motion associated with the first and second time points; and

assessing, by the one or more processors, facial liveliness of the subject by determining likelihood between the calculated expected pose of the face to the determined second pose of the face from the second facial image,

wherein assessing the facial liveliness of the subject comprises:

comparing a correlation between the determined second pose of the face and the calculated expected pose of the face to a predetermined threshold; and

determining whether the face of the subject is live based on a result of the comparison,

wherein assessing the facial liveliness of the subject comprises:

determining that the face of the subject is live in response to determining that the correlation is beyond the determined threshold.