Recognition of free-form gestures from orientation tracking of a handheld or wearable device
First Claim
1. A wearable device, comprising:
- a band to be worn on an arm of a user of the wearable device;
at least one sensor to generate first motion data, the at least one sensor carried by the wearable device, the wearable device attached to the band, the first motion data including a series of vectors indicative of a motion of the wearable device when the user performs a gesture with the arm, the first motion data associated with a first coordinate system;
an integrated circuit to execute software to;
transform the first motion data into second motion data, the second motion data associated with a second coordinate system different than the first coordinate system;
extract features from the second motion data, the features based on movement of at least one of a shoulder or an elbow; and
classify the motion of the wearable device by comparison of the features to a movement pattern; and
a display to present content, the content based on the classification of the motion of the wearable device.
2 Assignments
0 Petitions
Accused Products
Abstract
A user performs a gesture with a hand-held or wearable device capable of sensing its own orientation. Orientation data, in the form of a sequence of rotation vectors, is collected throughout the duration of the gesture. To construct a trace representing the shape of the gesture and the direction of device motion, the orientation data is processed by a robotic chain model with four or fewer degrees of freedom, simulating a set of joints moved by the user to perform the gesture (e.g., a shoulder and an elbow). To classify the gesture, a trace is compared to contents of a training database including many different users'"'"' versions of the gesture and analyzed by a learning module such as support vector machine.
17 Citations
17 Claims
-
1. A wearable device, comprising:
-
a band to be worn on an arm of a user of the wearable device; at least one sensor to generate first motion data, the at least one sensor carried by the wearable device, the wearable device attached to the band, the first motion data including a series of vectors indicative of a motion of the wearable device when the user performs a gesture with the arm, the first motion data associated with a first coordinate system; an integrated circuit to execute software to; transform the first motion data into second motion data, the second motion data associated with a second coordinate system different than the first coordinate system; extract features from the second motion data, the features based on movement of at least one of a shoulder or an elbow; and classify the motion of the wearable device by comparison of the features to a movement pattern; and a display to present content, the content based on the classification of the motion of the wearable device. - View Dependent Claims (2, 3, 4)
-
-
5. A wearable device, comprising:
-
a band to be worn on an arm of a user of the wearable device; at least one sensor to generate first motion data, the at least one sensor carried by the wearable device, the wearable device attached to the band, the first motion data including a series of vectors indicative of a motion of the wearable device when the user performs a gesture with the arm, the series of vectors including a sequence of rotation vectors, the first motion data associated with a first coordinate system; an integrated circuit to execute software to; convert the sequence of rotation vectors to a sequence of device positions using a robotic chain model; transform the first motion data into second motion data, the second motion data associated with a second coordinate system different than the first coordinate system; extract features from the second motion data; and classify the motion of the wearable device by comparison of the features to a movement pattern; and a display to present content, the content based on the classification of the motion of the wearable device. - View Dependent Claims (6, 7)
-
-
8. A machine-readable storage device comprising instructions that, when executed, cause a machine to at least:
-
generate first motion data via at least one sensor carried by a wearable device, the wearable device attached to a band to be worn on an arm of a user of the wearable device, the first motion data including a series of vectors indicative of a motion of the wearable device corresponding to a gesture with the arm, the first motion data associated with a first coordinate system; transform the first motion data into second motion data, the second motion data associated with a second coordinate system different than the first coordinate system; extract features from the second motion data, the features based on movement of at least one of a shoulder or an elbow; classify the motion of the wearable device by comparison of the features to a movement pattern; and present content based on the classification of the motion of the wearable device. - View Dependent Claims (9)
-
-
10. A machine-readable storage device comprising instructions that, when executed, cause a machine to at least:
-
generate first motion data via at least one sensor carried by a wearable device, the wearable device attached to a band to be worn on an arm of a user of the wearable device, the first motion data including a series of vectors indicative of a motion of the wearable device corresponding to a gesture with the arm, the series of vectors including a sequence of rotation vectors, the first motion data associated with a first coordinate system; convert the sequence of rotation vectors to a sequence of device positions using a robotic chain model; transform the first motion data into second motion data, the second motion data associated with a second coordinate system different than the first coordinate system; extract features from the second motion data; classify the motion of the wearable device by comparison of the features to a movement pattern; and present content based on the classification of the motion of the wearable device. - View Dependent Claims (11, 12)
-
-
13. A method comprising:
-
generating first motion data from an output of at least one sensor carried by a wearable device, the wearable device attached to a band to be worn on an arm of a user of the wearable device, the first motion data including a series of vectors indicative of a motion of the wearable device when the user performs a gesture with the arm, the first motion data associated with a first coordinate system; transforming the first motion data into second motion data, the second motion data associated with a second coordinate system different than the first coordinate system; extracting features from the second motion data, the features based on movement of at least one of a shoulder or an elbow; classifying the motion of the wearable device by comparison of the features to one or more movement patterns; and presenting content based on the classification of the motion of the wearable device. - View Dependent Claims (14)
-
-
15. A method comprising:
-
generating first motion data from an output of at least one sensor carried by a wearable device, the wearable device attached to a band to be worn on an arm of a user of the wearable device, the first motion data including a series of vectors indicative of a motion of the wearable device when the user performs a gesture with the arm, the series of vectors including a sequence of rotation vectors, the first motion data associated with a first coordinate system; converting the sequence of rotation vectors to a sequence of device positions using a robotic chain model; transforming the first motion data into second motion data, the second motion data associated with a second coordinate system different than the first coordinate system; extracting features from the second motion data; classifying the motion of the wearable device by comparison of the features to one or more movement patterns; and presenting content based on the classification of the motion of the wearable device. - View Dependent Claims (16, 17)
-
Specification