Tremor correction for gesture recognition

• US 10,429,935 B2
• Filed: 02/08/2016
• Issued: 10/01/2019
• Est. Priority Date: 02/08/2016
• Status: Active Grant

• Current Assignee: Comcast Cable Communications LLC (Comcast Corporation)
• Original Assignee: Comcast Cable Communications LLC (Comcast Corporation)
• Inventors: Hall, Jeffrey Allen, Monnerat, Edward David
• Primary Examiner(s): Sitta, Grant
• Application Number:
  US15/017,979
• Publication Number:
  US 20170228025A1
• Time in Patent Office: 1,331 Days
• Field of Search: 345156
• CPC Class Codes: G06F 3/017 Gesture based interaction, e.g. based on a set of recognized hand gestures interaction based on gestures traced on a digitiser G06F3/04883

Abstract

Motion detection computing devices may have difficulty determining precise motions of a user who suffers from unintended movement, such as tremors, associated with a physical or medical condition. Aspects described herein relate to motion compensation for detected motion input, e.g., 3D motion, from such users. As described herein, a system may detect the presence of rhythmic motion and/or other input characteristics within captured motion input from a user. The system may determine one or more filters to apply to the captured motion input based on the rhythmic motion, other input characteristics, and/or information in a user profile. The one or more filters may be applied to the captured motion input in order to suppress or minimize the effects of the rhythmic motion and/or other input characteristics thereby resulting in an enhanced user experience in an interactive system, such as a 3D, virtual and/or augmented reality environment.

Claims

1. A method comprising:

detecting, by a first computing device, a tremor in motion input obtained by one or more sensors;

determining, by the first computing device, whether to filter the obtained motion input, wherein the determining is based on;

one or more frequencies corresponding to predefined rhythmic gestures of valid user commands used by an application, and

a frequency of the detected tremor;

determining, by the first computing device, that a processing threshold has been satisfied, wherein the processing threshold is established based on a frequency of a rhythmic gesture command being higher than the frequency of the detected tremor;

sending, to a second computing device, a request for rhythmic filtering to suppress the detected tremor and generate a filtered motion input;

receiving, from the second computing device, the filtered motion input; and

transmitting, by the first computing device and based on a comparison of the filtered motion input with the one or more frequencies, one or more user commands to the application.

2. The method of claim 1, wherein detecting the tremor comprises:

determining that a pattern in the obtained motion input satisfies one or more predefined threshold values comprising at least a time value, a frequency value, or a combination thereof.

3. The method of claim 1, wherein the one or more user commands comprise commands to manipulate virtual reality objects in virtual three dimensional space.

4. The method of claim 1, further comprising:

determining, by the first computing device, an application type of the application, wherein the application is executed on the first computing device;

determining, based on the application type, a plurality of user commands for the application that are associated with rhythmic patterns; and

retrieving, by the first computing device, a plurality of frequencies associated with the plurality of user commands.

5. The method of claim 1, further comprising:

selecting a stored filter from a user profile; and

after determining that residual unintended rhythmic motion is present in the filtered motion input;

updating the stored filter; and

filtering the filtered motion input using the updated filter to generate a second filtered motion input.

6. The method of claim 1, wherein the obtained motion input comprises three dimensional position data associated with one or more tracked objects that are attached to one or more portions of a user'"'"'s body.

7. The method of claim 1, further comprising:

accessing a user profile to determine one or more medical conditions associated with a user;

determining, by the first computing device, a stored filter based on the one or more medical conditions;

filtering the filtered motion input using the stored filter to generate a second filtered motion input; and

transmitting, based on the second filtered motion input, a signal indicating one or more user commands.

14. The method of claim 1, further comprising:

determining, by the first computing device, a user;

selecting, by the first computing device and based on the user, a stored filter from a user profile; and

after determining that residual unintended rhythmic motion is not present in the filtered motion input;

reducing, based on a predefined value, a strength of the stored filter to generate a reduced filter, and

storing the reduced filter in the user profile.

8. A method comprising:

determining, by a first computing device, a first frequency corresponding to a predefined rhythmic gesture of a valid user command used by an application;

detecting, by the first computing device, a tremor in motion input obtained by one or more sensors;

determining, by the first computing device, that a processing threshold has been satisfied, wherein the processing threshold is established based on the predefined rhythmic gesture occurring at a higher frequency than the detected tremor;

sending, to a second computing device, a request for rhythmic filtering to suppress the detected tremor and generate a first filtered motion input; and

transmitting, by the first computing device and based on a comparison of the first filtered motion input with the first frequency, at least a first command to the application.

9. The method of claim 8, wherein the obtained motion input comprises an input user bias, the method further comprising:

determining, by the first computing device and based on the obtained motion input, an intended gesture;

detecting, by the first computing device and based on the intended gesture, the input user bias; and

filtering, by the first computing device, the obtained motion input to correct the input user bias.

10. The method of claim 8, wherein detecting the tremor further comprises:

determining, by the first computing device, a presence of the tremor based on a frequency of a periodic pattern in the obtained motion input satisfying a predefined temporal threshold and a predefined frequency threshold.

11. The method of claim 8,

wherein the processing threshold is further established based on a threshold number of users.

12. The method of claim 8, further comprising:

selecting, by the first computing device and based on user identification information, a stored filter;

updating the stored filter, and filtering the first filtered motion input using the updated filter to generate a second filtered motion input, wherein updating the stored filter is based on a determination that residual unintended rhythmic motion is present in the first filtered motion input; and

transmitting, based on the updated second filtered motion input, a signal indicating one or more user commands.

13. The method of claim 12, wherein selecting the stored filter based on the user identification information comprises:

accessing a user profile to determine one or more medical conditions and an age of a user; and

determining, by the first computing device and based on the one or more medical conditions and the age of the user, the stored filter.

15. The method of claim 8, wherein the obtained motion input comprises a horizontal gesture, the method further comprising:

determining, based on the obtained motion input, the horizontal gesture;

detecting unintended vertical movement within the obtained motion input; and

filtering the obtained motion input to correct the unintended vertical movement.

16. A system comprising:

a first computing device comprising; and

a second computing device;

wherein the first computing device comprises;

one or more first processors; and

memory storing first computer readable instructions that, when executed by the one or more first processors, cause the first computing device to;

detect a tremor in motion input obtained by one or more sensors;

determine whether to filter the obtained motion input, wherein the determining is based on;

one or more frequencies corresponding to predefined rhythmic gestures of valid user commands used by an application, and

a frequency of the detected tremor;

determine that a processing threshold has been satisfied, wherein the processing threshold is established based on a frequency of a rhythmic gesture command being higher than the frequency of the detected tremor;

send, to the second computing device, a request for rhythmic filtering; and

transmit, based on a comparison of a filtered motion input with the one or more frequencies, one or more user commands to the application;

wherein the second computing device comprises;

one or more second processors; and

memory storing second computer readable instructions that, when executed by the one or more second processors, cause the second computing device to;

filter, by using a first filter to suppress the detected tremor and generate the filtered motion input, the obtained motion input; and

transmit, to the first computing device, the filtered motion input.

17. The system of claim 16, wherein the second computer readable instructions, when executed by the one or more second processors, further cause the second computing device to:

select the first filter based on a user profile.

18. The system of claim 16, wherein the second computer readable instructions, when executed by the one or more second processors, further cause the second computing device to:

after determining that residual unintended rhythmic motion is present in the filtered motion input;

update the first filter; and

filter the filtered motion input using the updated first filter to generate a second filtered motion input.

19. The system of claim 16, wherein the second computer readable instructions, when executed by the one or more second processors, further cause the second computing device to:

select, based on a user profile, the first filter; and

after determining that residual unintended rhythmic motion is not present in the filtered motion input;

reduce, based on a predefined value, a strength of the first filter to generate a reduced filter, and

store the reduced filter in the user profile.