IP Library Granted Patent US 12669703
Granted Patent B2
US 12669703 · App. 18/857,365 · Granted Jun 30, 2026

Method for calibrating a head-mounted device

Inventors: Aurélie Le Cain (Creteil, FR); Amandine Debieuvre (Paris, FR); Bruno Amir (Saint-Maur-des-Foss, FR); Konogan Baranton (Montry, FR); Delphine Bernardin (Montreal, CA)
Assignee: Essilor International
G02B27/0093G02B27/017G02B2027/0178
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12669703
App. No.
18/857,365
Granted
Jun 30, 2026
Kind
B2
Abstract

The invention relates to method for calibrating a head-mounted device ( 1 ), the head-mounted device ( 1 ) comprising a frame ( 2 ) and at least one sensor ( 4 ) mounted on the frame ( 2 ), for measuring position and/or motion, the method being implemented while the head-mounted device ( 1 ) is worn on a wearer's head, the method comprising steps of: —detection (DET) of a known repetitive movement pattern of the wearer's head by the sensor ( 4 ), —acquisition (ACQ) of three-dimensional motion data by the sensor ( 4 ), —selection (SEL) of data relative to the movement pattern from the motion data, —numerical treatment (NUM) of the selected data to obtain a reference base relative to the wearer's head, and —calibrating (CAL) the sensor ( 4 ) based on the obtained reference base.

Claims (21)

1 . A method for calibrating a head-mounted device including a frame and at least one sensor mounted on the frame and configured to measure one or more of position and motion, the method being implemented while the head-mounted device is worn on a head of a wearer, the method comprising:

detecting a known repetitive movement pattern of the head of the wearer by the sensor;

acquiring three-dimensional motion data by the sensor;

selecting data relative to the known repetitive movement pattern from the three-dimensional motion data;

numerically treating the selected data to obtain a reference base relative to the head of the wearer, by detecting and localizing at least one head reference feature through a numerical treatment tool applied to the three-dimensional motion data, the numerical treatment tool performing at least a principal components analysis to obtain eigenvectors from the selected data, the at least one head reference feature comprising at least one of a sagittal plane, a frontal plane, and a transverse plane, relative to the head of the wearer, the reference base being determined based on the localization of said at least one reference feature; and

calibrating the sensor based on the obtained reference base.

2 . The method according to claim 1 , wherein the movement pattern is caused by the wearer walking while wearing the head-mounted device, the detecting comprising detecting a walking movement pattern and the selecting comprising selecting data relative to the walking movement pattern.

3 . The method according to claim 2 , wherein, during the selecting, data corresponding to a predetermined number of first walking steps of the wearer is discarded.

4 . The method according to claim 1 , further comprising preliminarily triggering a request to recalibrate the head-mounted device, subsequent operations of the method being implemented only when the request to recalibrate is issued.

5 . The method according to claim 4 , wherein the request to recalibrate the head-mounted device is issued when the device is put on the face of the wearer after having been taken off.

6 . The method according to claim 4 , wherein the request to recalibrate the head-mounted device is issued when a predetermined recalibration period expires.

7 . The method according to claim 6 , wherein said recalibration period is calculated based on pre registered characteristics of the wearer.

8 . The method according to claim 6 , wherein said recalibration period is adjusted based on detected or scheduled activities of the wearer while wearing the head-mounted device.

9 . The method according to claim 1 , wherein the numerical treatment involves at least one of a signal processing, a minimum variance calculation, a principal components analysis, a machine learning algorithm, and a support vector analysis.

10 . A method for adjusting a posture of a wearer of a head-mounted device comprising:

calibrating the head-mounted device through the method according to claim 1 ;

using data obtained for the sensor, calibrated in the obtained reference base, to determine a current posture of the wearer;

determining one or more of a correct posture for the wearer and a correct acceleration of the head of the wearer, relative to a detected activity of the wearer;

comparing the current posture to the correct posture to assess a correctness; and

prompting the wearer to adjust said current posture when said current posture is incorrect.

11 . A non-transitory computer-readable storage medium comprising instructions for executing a computer program which, when executed by a controller of a head mounted device including a frame and at least one sensor mounted on the frame configured to measure one of position and motion, or connected to said head-mounted device, enables implementation of the calibration method according to claim 1 .