Diffusion-based handedness classification for touch-based input
Methods, systems, apparatuses, and computer program products are provided herein for determining the handedness of input provided by a user via a touch interface. For instance, for each touch-based input detected, a score indicating a probability whether the touch-based input was inputted by a particular hand of the user is generated. A classification for the touch-based input is then generated based on a drift diffusion model-based technique in which inter-dependencies between a series of touch-based input are approximated. The determined classifications are aggregated and compared to threshold(s) to determine the handedness of the user. For example, if the aggregated classifications meet a first threshold, then a determination is made that touch-based input provided by a user was inputted by the user's left hand. If the aggregated classifications meet a second threshold, then a determination is made that touch-based input provided by the user was inputted by the user's right hand.
1. A computing device, comprising:
at least one processor circuit; and
at least one memory that stores program code configured to be executed by the at least one processor circuit, the program code comprising:
a handedness classifier configured to:
receive, by a touch interface of the computing device, a first touch-based input applied by a contact instrument;
generate a score indicating a probability whether the first touch-based input was inputted by a particular hand of a user;
generate a first classification for the first touch-based input based at least on the score, a second classification generated for a previously-received touch-based input, and a diffusion factor representative of a dependency between the first touch-based input and the previously-received touch-based input;
determine whether the first classification meets a first predetermined threshold or a second predetermined threshold;
in response to a determination that the first classification meets the first predetermined threshold, classify the first touch-based input as being inputted from the left hand of the user; and
in response to determination that the first classification meets the second predetermined threshold, classify the first touch-based input as being inputted from the right hand of the user.
2. The computing device of claim 1 , wherein the handedness classifier is further configured to:
update a configuration setting of at least one of an operating system executing on the computing device or an application executing on the computing device; and
receive, by the touch interface, a second touch-based input applied by the contact instrument, the second touch-based input causing a graphical user interface element to be displayed in accordance with the configuration setting, the configuration setting causing the graphical user interface element to be displayed a manner such that the graphical user interface element is unobscured by the particular hand of the user by which the second touch-based input is provided.
3. The computing device of claim 2 , wherein the handedness classifier is further configured to:
reject a third touch-based input, received by the touch interface, corresponding to a touch blob, the touch blob corresponding to a portion of the particular hand of the user that is applied to the touch interface while the second touch-based input is received.
4. The computing device of claim 2 , the second touch-based input is received in accordance with at least one of:
a time period that exceeds a third predetermined threshold; or
at a pressure that exceeds a fourth predetermined threshold.
5. The computing device of claim 1 , wherein the score is generated based at least on one feature of a plurality of features, the plurality of features comprising:
a first location of the touch interface at which the first touch-based input was applied by the contact instrument;
a second location of the touch interface at which a touch blob was detected by the touch interface, the touch blob corresponding to a portion of the particular hand of the user that is applied to the touch interface while the first touch-based input is received;
an angle at which the contact instrument is positioned with respect to the first location and the second location;
a tilt angle of the contact instrument with respect to the touch interface; and
an azimuth of the contact instrument.
6. The computing device of claim 5 , wherein the handedness classifier is further configured to:
generate a feature vector based on the plurality of features; and
provide the feature vector as an input to a machine learning model that generates the score.
7. The computing device of claim 1 , wherein the handedness classifier is further configured to:
determine a second score indicating that the first touch-based input was inputted by another particular hand of the user;
generate a first value based on the first score and the second score;
scale the first value by the diffusion factor to generate a second value; and
combine the second value with the second classification generated for the previously-received touch-based input to generate the first classification.
8. A method implemented on a computing device, comprising:
receiving, by a touch interface of the computing device, a first touch-based input applied by a contact instrument;
generating a score indicating a probability whether the first touch-based input was inputted by a particular hand of a user;
generating a first classification for the first touch-based input based at least on the score, a second classification generated for a previously-received touch-based input, and a diffusion factor representative of a dependency between the first touch-based input and the previously-received touch-based input;
determining whether the first classification meets a first predetermined threshold or a second predetermined threshold;
in response to determining that the first classification meets the first predetermined threshold, classifying the first touch-based input as being inputted from the left hand of the user; and
in response to determining that the first classification meets the second predetermined threshold, classifying the first touch-based input as being inputted from the right hand of the user.
9. The method of claim 8 , further comprising:
updating a configuration setting of at least one of an operating system executing on the computing device or an application executing on the computing device;
receiving, by the touch interface, a second touch-based input applied by the contact instrument; and
responsive to receiving the second touch-based input, displaying a graphical user interface element in accordance with the configuration setting, the configuration setting causing the graphical user interface element to be displayed a manner such that the graphical user interface element is unobscured by the particular hand of the user by which the second touch-based input is provided.
10. The method of claim 9 , further comprising:
rejecting a third touch-based input, received by the touch interface, corresponding to a touch blob, the touch blob corresponding to a portion of the particular hand of the user that is applied to the touch interface while the second touch-based input is received.
11. The method of claim 9 , wherein the second touch-based input is received in accordance with at least one of:
a time period that exceeds a third predetermined threshold; or
at a pressure that exceeds a fourth predetermined threshold.
12. The method of claim 8 , wherein the score is generated based at least on one feature of a plurality of features, the plurality of features comprising:
a first location of the touch interface at which the first touch-based input was applied by the contact instrument;
a second location of the touch interface at which a touch blob was detected by the touch interface, the touch blob corresponding to a portion of the particular hand of the user that is applied to the touch interface while the first touch-based input is received;
an angle at which the contact instrument is positioned with respect to the first location and the second location;
a tilt angle of the contact instrument with respect to the touch interface; and
an azimuth of the contact instrument.
13. The method of claim 12 , wherein said generating the score comprises:
generating a feature vector based on the plurality of features; and
providing the feature vector as an input to a machine learning model that generates the score.
14. The method of claim 8 , wherein said generating the first classification comprises:
determining a second score indicating that the first touch-based input was inputted by another particular hand of the user;
generating a first value based on the first score and the second score;
scaling the first value by the diffusion factor to generate a second value; and
combining the second value with the second classification generated for the previously-received touch-based input to generate the first classification.
15. A computer-readable storage medium having program instructions recorded thereon that, when executed by at least one processor of a computing device, perform a method, the method comprising:
receiving, by a touch interface of the computing device, a first touch-based input applied by a contact instrument;
generating a score indicating a probability whether the first touch-based input was inputted by a particular hand of a user;
generating a first classification for the first touch-based input based at least on the score, a second classification generated for a previously-received touch-based input, and a diffusion factor representative of a dependency between the first touch-based input and the previously-received touch-based input;
determining whether the first classification meets a first predetermined threshold or a second predetermined threshold;
in response to determining that the first classification meets the first predetermined threshold, classifying the first touch-based input as being inputted from the left hand of the user; and
in response to determining that the first classification meets the second predetermined threshold, classifying the first touch-based input as being inputted from the right hand of the user.
16. The computer-readable storage medium of claim 15 , the method further comprising:
updating a configuration setting of at least one of an operating system executing on the computing device or an application executing on the computing device;
receiving, by the touch interface, a second touch-based input applied by the contact instrument; and
responsive to receiving the second touch-based input, displaying a graphical user interface element in accordance with the configuration setting, the configuration setting causing the graphical user interface element to be displayed a manner such that the graphical user interface element is unobscured by the particular hand of the user by which the second touch-based input is provided.
17. The computer-readable storage medium of claim 16 , the method further comprising:
rejecting a third touch-based input, received by the touch interface, corresponding to a touch blob, the touch blob corresponding to a portion of the particular hand of the user that is applied to the touch interface while the second touch-based input is received.
18. The computer-readable storage medium of claim 16 , wherein the second touch-based input is received in accordance with at least one of:
a time period that exceeds a third predetermined threshold; or
at a pressure that exceeds a fourth predetermined threshold.
19. The computer-readable storage medium of claim 15 , wherein the score is generated based at least on one feature of a plurality of features, the plurality of features comprising:
a first location of the touch interface at which the first touch-based input was applied by the contact instrument;
a second location of the touch interface at which a touch blob was detected by the touch interface, the touch blob corresponding to a portion of the particular hand of the user that is applied to the touch interface while the first touch-based input is received;
an angle at which the contact instrument is positioned with respect to the first location and the second location;
a tilt angle of the contact instrument with respect to the touch interface; and
an azimuth of the contact instrument.
20. The computer-readable storage medium of claim 19 , wherein said generating the score comprises:
generating a feature vector based on the plurality of features; and
providing the feature vector as an input to a machine learning model that generates the score.