Driving method for calculating interpupillary distance and related head-mounted device

1. A driving method, suitable for a head mounted device (HMD), comprising:

moving a first image capture unit and a second image capture unit of the HMD to respectively capture two left-eye images and two right-eye images;

calculating a first eye relief according to at least one left-eye feature in the two left-eye images;

calculating a second eye relief according to at least one right-eye feature in the two right-eye images;

calculating an interpupillary distance (IPD) according to the first eye relief and the second eye relief; and

adjusting, according to the IPD, a distance between a first lens and a second lens of the HMD.

2. The driving method of claim 1 , wherein the operation of calculating the first eye relief comprises:

calculating a left-eye disparity of the at least one left-eye feature between the two left-eye images; and

calculating the first eye relief according to a plurality of camera parameters related to the first image capture unit or the second image capture unit, a displacement of the first image capture unit, and the left-eye disparity.

3. The driving method of claim 1 , wherein the operation of calculating the IPD comprises:

calculating a first distance between the at least one left-eye feature and an optical axis of the first image capture unit;

calculating a second distance between the at least one right-eye feature and an optical axis of the second image capture unit;

calculating, according to the first eye relief, the first distance, and a plurality of camera parameters related to the first image capture unit or the second image capture unit, a third distance between the left pupil and the optical axis of the first image capture unit;

calculating, according to the second eye relief, the second distance, and the plurality of camera parameters, a fourth distance between the right pupil and the optical axis of the second image capture unit; and

calculating the IPD according to the third distance, the fourth distance, and a distance between the first image capture unit and the second image capture unit.

4. The driving method of claim 3 , wherein the first display is moved along with the first image capture unit, the second display is moved along with the second image capture unit,

when the first display is moved in a first direction, the first display shows a first interest point having a moving direction opposite to the first direction,

when the second display is moved in a second direction, the second display shows a second interest point having a moving direction opposite to the second direction.

5. The driving method of claim 1 , wherein the operation of calculating the first eye relief comprises:

recognizing, from the two left-eye images, a plurality of left-eye features comprising the at least one left-eye feature;

calculating, according to the plurality of left-eye features, a first predicted coordinate and a second predicted coordinate of the left pupil respectively in the two left-eye images and both corresponding to a situation that a user is substantially staring straight forward;

calculating a left-eye disparity between the first predicted coordinate and the second predicted coordinate; and

calculating the first eye relief according to a plurality of camera parameters related to the first image capture unit or the second image capture unit, a displacement of the first image capture unit, and the left-eye disparity.

6. The driving method of claim 5 , wherein the operation of calculating the second eye relief comprises:

recognizing, from the two right-eye images, a plurality of right-eye features comprising the at least one right-eye feature;

calculating, according to the plurality of right-eye features, a third predicted coordinate and a fourth predicted coordinate of the right pupil respectively in the two right-eye images and both corresponding to the situation that the user is substantially staring straight forward;

calculating a right-eye disparity between the third predicted coordinate and the fourth predicted coordinate; and

calculating the second eye relief according to the plurality of camera parameters, a displacement of the second image capture unit, and the right-eye disparity.

7. The driving method of claim 6 , wherein the operation of calculating the IPD comprises:

calculating a first distance between the second predicted coordinate and an optical axis of the first image capture unit;

calculating a second distance between the fourth predicted coordinate and an optical axis of the second image capture unit;

calculating, according to the first eye relief, the first distance, and the plurality of camera parameters, a third distance between the optical axis of the first image capture unit and a predicted left-pupil position in an accommodating space of the HMD;

calculating, according to the second eye relief, the second distance, and the plurality of camera parameters, a fourth distance between the optical axis of the second image capture unit and a predicted right-pupil position in the accommodating space; and

calculating the IPD according to the third distance, the fourth distance, and a distance between the first image capture unit and the second image capture unit.

8. The driving method of claim 6 , further comprising:

constructing and storing, according to the plurality of left-eye features and the plurality of right-eye features, a 3-dimensional (3D) left-eye model and a 3D right-eye model located in an accommodating space of the HMD; and

automatically adjust or prompting the user to manually adjust, according to the 3D left-eye model and the 3D right-eye model, the distance between the first lens and the second lens.

9. The driving method of claim 6 , wherein the driving method further comprises:

utilizing the first image capture unit to capture a plurality of left-eye images;

utilizing the second image capture unit to capture a plurality of right-eye images; and

recognizing the plurality of left-eye features from the plurality of left-eye images and recognizing the plurality of right-eye features from the plurality of right-eye images to determine a point of gaze of the user,

wherein the point of gaze is configured to be used in rendering pictures provided to the first display and the second display.

10. The driving method of claim 1 , wherein when adjusts the distance between the first lens and the second lens, automatically adjusts or prompts the user to manually adjust the distance between a center of the first lens and a center of the second lens to be equal to the IPD.

11. A head-mounted device (HMD), comprising:

a first lens;

a second lens;

a first image capture unit, disposed at a first housing with the first lens;

a second image capture unit, disposed at a second housing with the second lens; and

a processing circuit, being adapted to:

move the first image capture unit and the second image capture unit to respectively capture two left-eye images and two right-eye images;

calculate, according to at least one left-eye feature in the two left-eye images, a first eye relief between the first image capture unit and a left pupil;

calculate, according to at least one right-eye feature in the two right-eye images, a second eye relief between the second image capture unit and a right pupil;

calculate an interpupillary distance (IPD) according to the first eye relief and the second eye relief; and

automatically adjust or prompt the user to manually adjust, according to the IPD, a distance between the first lens and the second lens.

12. The HMD of claim 11 , wherein when calculating the first eye relief, the processing circuit is further adapted to:

calculate a left-eye disparity of the at least one left-eye feature between the two left-eye images; and

calculate the first eye relief according to a plurality of camera parameters related to the first image capture unit or the second image capture unit, a displacement of the first image capture unit, and the left-eye disparity.

13. The HMD of claim 11 , wherein when calculating the IPD, the processing circuit is further adapted to:

calculate a first distance between the at least one left-eye feature and an optical axis of the first image capture unit;

calculate a second distance between the at least one right-eye feature and an optical axis of the second image capture unit;

calculate, according to the first eye relief, the first distance, and a plurality of camera parameters related to the first image capture unit or the second image capture unit, a third distance between the left pupil and the optical axis of the first image capture unit;

calculate, according to the second eye relief, the second distance, and the plurality of camera parameters, a fourth distance between the right pupil and the optical axis of the second image capture unit; and

calculate the IPD according to the third distance, the fourth distance, and a distance between the first image capture unit and the second image capture unit.

14. The HMD of claim 13 , further comprising:

a first display, disposed at the first housing; and

a second display, disposed at the second housing,

wherein the first display is moved along with the first image capture unit, the second display is moved along with the second image capture unit,

when the first display is moved in a first direction, the first display shows a first interest point having a moving direction opposite to the first direction,

when the second display is moved in a second direction, the second display shows a second interest point having a moving direction opposite to the second direction.

15. The HMD of claim 11 , wherein when calculating the first eye relief, the processing circuit is further adapted to:

recognize, from the two left-eye images, a plurality of left-eye features comprising the at least one left-eye feature;

calculate, according to the plurality of left-eye features, a first predicted coordinate and a second predicted coordinate of the left pupil respectively in the two left-eye images and both corresponding to a situation that a user is substantially staring straight forward;

calculate a left-eye disparity between the first predicted coordinate and the second predicted coordinate; and

calculate the first eye relief according to a plurality of camera parameters related to the first image capture unit or the second image capture unit, a displacement of the first image capture unit, and the left-eye disparity.

16. The HMD of claim 15 , wherein when calculating the second eye relief, the processing circuit is further adapted to:

recognize, from the two right-eye images, a plurality of right-eye features comprising the at least one right-eye feature;

calculate, according to the plurality of right-eye features, a third predicted coordinate and a fourth predicted coordinate of the right pupil respectively in the two right-eye images and both corresponding to the situation that the user is substantially staring straight forward;

calculate a right-eye disparity between the third predicted coordinate and the fourth predicted coordinate; and

calculate the second eye relief according to the plurality of camera parameters, a displacement of the second image capture unit, and the right-eye disparity.

17. The HMD of claim 16 , wherein when calculating the IPD, the processing circuit is further adapted to:

calculate a first distance between the second predicted coordinate and an optical axis of the first image capture unit;

calculate a second distance between the fourth predicted coordinate and an optical axis of the second image capture unit;

calculate, according to the first eye relief, the first distance, and the plurality of camera parameters, a third distance between the optical axis of the first image capture unit and a predicted left-pupil position in an accommodating space of the HMD;

calculate, according to the second eye relief, the second distance, and the plurality of camera parameters, a fourth distance between the optical axis of the second image capture unit and a predicted right-pupil position in the accommodating space; and

calculate the IPD according to the third distance, the fourth distance, and a distance between the first image capture unit and the second image capture unit.

18. The HMD of claim 16 , wherein the processing circuit is further adapted to:

construct and store, according to the plurality of left-eye features and the plurality of right-eye features, a 3D left-eye model and a 3D right-eye model located in an accommodating space of the HMD; and

automatically adjust or prompt the user to manually adjust, according to the 3D left-eye model and the 3D right-eye model, the distance between the first lens and the second lens.

19. The HMD of claim 16 , the processing circuit is further adapted to:

instruct the first image capture unit to capture a plurality of left-eye images;

instruct the second image capture unit to capture a plurality of right-eye images; and

recognize the plurality of left-eye features from the plurality of left-eye images and recognize the plurality of right-eye features from the plurality of right-eye images to determine a point of gaze of the user,

wherein the point of gaze is configured to be used in rendering pictures provided to the first display and the second display.

20. The HMD of claim 11 , wherein the distance between a center of the first lens and a center of the second lens is adjusted to be equal to the IPD.