Orientation determination based on both images and inertial measurement units
A system to track orientations of parts of a user based on both images and inertial measurement units (IMUs). For example, the system receives images showing a portion of the user wearing sensor modules. The system receives a first set of orientation measurements generated by the sensor modules attached to some parts of the user. The system determines the second set of orientation measurements of one or more features of the portion of the user from the images. The system provides the first set of orientation measurements and the second set of orientation measurements as input to an artificial neural network that is configured to predict orientation measurements of the one or more other parts of the user that would be measured by additional sensor modules if the additional sensor modules were to be attached to the other parts of the user.
1. A method, comprising:
receiving an image from a camera showing a portion of a user wearing a plurality of sensor modules, each of the sensor modules having an inertial measurement unit configured to determine an orientation of the inertial measurement unit, wherein a skeleton model of the user includes a kinematic chain having a plurality of rigid parts connected via joints, wherein orientations of first rigid parts in the kinematic chain are tracked via the sensor modules worn on the user;
wherein one or more second rigid parts in the kinematic chain are not tracked using inertial measurement units;
receiving first orientation measurements generated by the sensor modules for the first rigid parts;
determining second orientation measurements of one or more features of the portion of the user from the image using a second artificial neural network;
providing the first orientation measurements and the second orientation measurements as input to a first artificial neural network;
generating, by the first artificial neural network, predicted orientation measurements of the one or more second rigid parts based on the first orientation measurements and the second orientation measurements; and
periodically capturing images using the camera, the camera being in a head mounted display, to provide input to the second artificial neural network, the second artificial neural network generating, from the images, orientation measurements of one or more features as input for the first artificial neural network at a frequency lower than a frequency of the sensor modules generating orientation measurements of the first rigid parts as input for the first artificial neural network.
2. The method of claim 1 , wherein the second artificial neural network is a convolutional neural network.
3. The method of claim 2 , wherein the first artificial neural network is a recurrent neural network.
4. The method of claim 1 , wherein the first artificial neural network is trained to predict orientation measurements of the second rigid parts measured using an optical tracking system.
5. The method of claim 4 , wherein the first artificial neural network is trained to predict orientation measurements of the first rigid parts measured using the optical tracking system.
6. A system, comprising:
one or more processors; and
a non-transitory computer-readable medium including one or more sequences of instructions that, when executed by the one or more processors, cause:
receiving an image from a camera showing a portion of a user wearing a plurality of sensor modules, each of the sensor modules having an inertial measurement unit configured to determine an orientation of the inertial measurement unit, wherein a skeleton model of the user includes a kinematic chain having a plurality of rigid parts connected via joints, wherein orientations of first rigid parts in the kinematic chain are tracked via the sensor modules worn on the user;
wherein one or more second rigid parts in the kinematic chain are not tracked using inertial measurement units;
receiving first orientation measurements generated by the sensor modules for the first rigid parts;
determining second orientation measurements of one or more features of the portion of the user from the image using a second artificial neural network;
providing the first orientation measurements and the second orientation measurements as input to a first artificial neural network;
generating, by the first artificial neural network, predicted orientation measurements of the one or more second rigid parts based on the first orientation measurements and the second orientation measurements; and
periodically capturing images using the camera, the camera being in a head mounted display, to provide input to the second artificial neural network, the second artificial neural network generating, from the images, orientation measurements of one or more features as input for the first artificial neural network at a frequency lower than a frequency of the sensor modules generating orientation measurements of the first rigid parts as input for the first artificial neural network.
7. The system of claim 6 , wherein the second artificial neural network is a convolutional neural network.
8. The system of claim 7 , wherein the first artificial neural network is a recurrent neural network.
9. The system of claim 6 , wherein the first artificial neural network is trained to predict orientation measurements of the second rigid parts measured using an optical tracking system.
10. The system of claim 9 , wherein the first artificial neural network is trained to predict orientation measurements of the first rigid parts measured using the optical tracking system.
11. A non-transitory computer storage medium storing instructions which, when executed by a computing device, cause the computing device to perform a method, the method comprising:
receiving an image from a camera showing a portion of a user wearing a plurality of sensor modules, each of the sensor modules having an inertial measurement unit configured to determine an orientation of the inertial measurement unit, wherein a skeleton model of the user includes a kinematic chain having a plurality of rigid parts connected via joints, wherein orientations of first rigid parts in the kinematic chain are tracked via the sensor modules worn on the user;
wherein one or more second rigid parts in the kinematic chain are not tracked using inertial measurement units;
receiving first orientation measurements generated by the sensor modules for the first rigid parts;
determining second orientation measurements of one or more features of the portion of the user from the image using a second artificial neural network;
providing the first orientation measurements and the second orientation measurements as input to a first artificial neural network;
generating, by the first artificial neural network, predicted orientation measurements of the one or more second rigid parts based on the first orientation measurements and the second orientation measurements; and
periodically capturing images using the camera, the camera being in a head mounted display, to provide input to the second artificial neural network, the second artificial neural network generating, from the images, orientation measurements of one or more features as input for the first artificial neural network at a frequency lower than a frequency of the sensor modules generating orientation measurements of the first rigid parts as input for the first artificial neural network.
12. The non-transitory computer storage medium of claim 11 , wherein the second artificial neural network is a convolutional neural network.
13. The non-transitory computer storage medium of claim 12 , wherein the first artificial neural network is a recurrent neural network.
14. The non-transitory computer storage medium of claim 11 , wherein the first artificial neural network is trained to predict orientation measurements of the second rigid parts measured using an optical tracking system.