Creating personalized athletic-performance models

• US 10,010,753 B2
• Filed: 12/30/2015
• Issued: 07/03/2018
• Est. Priority Date: 12/30/2015
• Status: Expired due to Fees

• Current Assignee: IAM Sports & Entertainment
• Original Assignee: IAM Sports & Entertainment
• Inventors: Brothers, Raymond Joseph, Gdoutos, Eleftherios E., Mulliken, Adam, Pye, John
• Primary Examiner(s): DUFFY, DAVID W
• Application Number:
  US14/985,111
• Publication Number:
  US 20170189756A1
• Time in Patent Office: 916 Days
• CPC Class Codes: A63B 2024/0012 Comparing movements or motion sequences with a registered reference
• CPC Class Codes: A63B 2209/10 with adhesive type surfaces, i.e. hook and loop-type fastener
• CPC Class Codes: A63B 2220/12 Absolute positions, e.g. by using GPS
• CPC Class Codes: A63B 2220/13 Relative positions
• CPC Class Codes: A63B 2220/24 Angular displacement
• CPC Class Codes: A63B 2220/40 Acceleration
• CPC Class Codes: A63B 2220/53 Force of an impact, e.g. blow or punch
• CPC Class Codes: A63B 2220/803 Motion sensors
• CPC Class Codes: A63B 2220/805 Optical or opto-electronic sensors
• CPC Class Codes: A63B 2220/806 Video cameras
• CPC Class Codes: A63B 2220/836 Sensors arranged on the body of the user
• CPC Class Codes: A63B 2225/50 Wireless data transmission, e.g. by radio transmitters or telemetry
• CPC Class Codes: A63B 2230/10 electroencephalographic signals
• CPC Class Codes: A63B 2230/60 muscle strain, i.e. measured on the user
• CPC Class Codes: A63B 2243/0037 Basketball
• CPC Class Codes: A63B 24/0006 Computerised comparison for qualitative assessment of motion sequences or the course of a movement
• CPC Class Codes: A63B 24/0075 Means for generating exercise programs or schemes, e.g. computerized virtual trainer, e.g. using expert databases
• CPC Class Codes: A63B 43/008 with means for improving visibility, e.g. special markings or colours illuminating or reflecting means A63B43/06
• CPC Class Codes: A63B 69/0071 for basketball
• CPC Class Codes: A63B 69/3623 for driving A63B69/3608, A63B69/3658, A63B69/3661, A63B69/3667 and A63B69/3691 take precedence
• CPC Class Codes: G06F 18/22 : Matching criteria, e.g. proximity measures
• CPC Class Codes: G06V 10/245 : by locating a pattern; Special marks for positioning
• CPC Class Codes: G06V 20/20 : in augmented reality scenes
• CPC Class Codes: G06V 40/23 : Recognition of whole body movements, e.g. for sport training
• CPC Class Codes: G06V 40/28 : Recognition of hand or arm movements, e.g. recognition of deaf sign language static hand signs G06V40/113

Abstract

In one embodiment, a method includes receiving, from one or more sensors, biomechanical data of a first user performing a plurality of actions of a first action-type and outcome data of a plurality of outcomes corresponding to the respective plurality of actions. A plurality of sets of action-parameter values may be determined based on the received biomechanical data, where each of the sets of action-parameter values corresponds to a respective action of the plurality of actions. An athletic-performance model of the first user may be generated based on the plurality of sets of action-parameter values. The athletic-performance model may be a statistical model including probabilities computed with respect to the sets of action-parameter values and the outcome data.

Claims

1. A method comprising, by one or more computing devices:

receiving, from one or more sensors comprising a point-of-view camera positioned on a first user, first biomechanical data of the first user performing a first plurality of actions of a first action-type and first outcome data comprising a first plurality of outcomes of the first user, wherein each action of the first plurality of actions is associated with a respective outcome of the first plurality of outcomes;

determining, based on the received first biomechanical data, a first plurality of sets of action-parameter values for respective action parameters of a plurality of action parameters, each of the first plurality of sets of action-parameter values corresponding to an action of the first plurality of actions, the first plurality of sets of action-parameter values being associated with respective outcomes of the first plurality of outcomes;

ranking the plurality of action parameters for the first user, wherein rankings are determined based on a first analysis of the first plurality of sets of action-parameter values and of the respective outcomes of the first plurality of outcomes;

assigning weights to the plurality of action parameters based on the rankings; and

generating a first athletic-performance model of the first user based on the first plurality of sets of action-parameter values, on the assigned weights of the plurality of action parameters, and on the first outcome data, wherein the first athletic-performance model comprises target ranges of action-parameter values computed for the first user based on a second analysis of the first plurality of sets of action-parameter values, of the assigned weights of the plurality of action parameters, and of the first outcome data.

2. The method of claim 1, further comprising:

receiving, from the one or more sensors in real time, real-time biomechanical data of the first user performing a real-time action of the first action-type and real-time outcome data of an outcome corresponding to the real-time action;

determining, based on the received real-time biomechanical data, an updated set of action-parameter values for respective action parameters of the plurality of action parameters; and

dynamically updating the first athletic-performance model based on the updated set of action-parameter values, on the assigned weights of the plurality of action parameters, and on the real-time outcome data to generate an updated athletic-performance model for the first user.

3. The method of claim 2, wherein dynamically updating the first athletic-performance model comprises comparing the updated set of action-parameter values to the first plurality of sets of action-parameter values.

4. The method of claim 2, wherein dynamically updating the first athletic-performance model comprises determining an updated current skill level of the first user.

5. The method of claim 4, wherein the updated current skill level is determined based on one or more measures of variance computed with respect to the first plurality of sets of action-parameter values and the updated set of action-parameter values.

6. The method of claim 2, further comprising generating a report of athletic-performance feedback for the first user based on the updated athletic-performance model.

7. The method of claim 6, wherein the report is further based on a comparison of the first athletic-performance model to the updated athletic-performance model.

8. The method of claim 2, further comprising generating a second athletic-performance model of a second user based on a second plurality of sets of action-parameter values of the second user, the second plurality of sets of action-parameter values being determined based on second biomechanical data of the second user performing a second plurality of actions of the first action-type and on second outcome data comprising a second plurality of outcomes corresponding to the second plurality of actions.

9. The method of claim 8, further comprising:

generating a report of athletic-performance feedback for the first user based on a comparison of the updated athletic-performance model to the second athletic-performance model of the second user.

10. The method of claim 9, wherein the comparison comprises comparing the updated set of action-parameter values and the first plurality of sets of action-parameter values of the first user to the second plurality of sets of action-parameter values of the second user.

11. The method of claim 2, further comprising generating an n-dimensional plot based on the updated athletic-performance model, wherein the n-dimensional plot comprises n dimensions, and wherein each of the n dimensions corresponds to an action parameter of the plurality of action parameters.

12. The method of claim 11, wherein the n-dimensional plot comprises a plurality of probability contours, and wherein each probability contour corresponds to a respective target range of the target ranges of action-parameter values, each probability contour representing a likelihood of a positive outcome for the action-parameter values corresponding to the respective target range.

13. The method of claim 1, wherein the first action-type is a basketball shot.

14. The method of claim 13, wherein the plurality of action parameters comprises a release angle of a basketball, the basketball being used by the first user in performing an action of the basketball shot.

15. The method of claim 14, wherein the plurality of action parameters comprises a release velocity of the basketball.

16. One or more computer-readable non-transitory storage media embodying software that is operable when executed to:

receive, from one or more sensors comprising a point-of-view camera positioned on a first user, first biomechanical data of the first user performing a first plurality of actions of a first action-type and first outcome data comprising a first plurality of outcomes of the first user, wherein each action of the first plurality of actions is associated with a respective outcome of the first plurality of outcomes;

determine, based on the received first biomechanical data, a first plurality of sets of action-parameter values for respective action parameters of a plurality of action parameters, each of the first plurality of sets of action-parameter values corresponding to an action of the first plurality of actions, the first plurality of sets of action-parameter values being associated with respective outcomes of the first plurality of outcomes;

rank the plurality of action parameters for the first user, wherein rankings are determined based on a first analysis of the first plurality of sets of action-parameter values and of the respective outcomes of the first plurality of outcomes;

assign weights to the plurality of action parameters based on the rankings; and

generate a first athletic-performance model of the first user based on the first plurality of sets of action-parameter values, on the assigned weights of the plurality of action parameters, and on the first outcome data, wherein the first athletic-performance model comprises target ranges of action-parameter values computed for the first user based on a second analysis of the first plurality of sets of action-parameter values, of the assigned weights of the plurality of action parameters, and of the first outcome data.

17. The media of claim 16, wherein the software is further operable when executed to:

receive, from the one or more sensors in real time, real-time biomechanical data of the first user performing a real-time action of the first action-type and real-time outcome data of an outcome corresponding to the real-time action;

determine, based on the received real-time biomechanical data, an updated set of action-parameter values; and

dynamically update the first athletic-performance model based on the updated set of action-parameter values.

18. A system comprising:

one or more processors; and

a memory coupled to the one or more processors comprising instructions executable by the one or more processors, the one or more processors operable when executing the instructions to;

receive, from one or more sensors comprising a point-of-view camera positioned on a first user, first biomechanical data of the first user performing a first plurality of actions of a first action-type and first outcome data comprising a first plurality of outcomes of the first user, wherein each action of the first plurality of actions is associated with a respective outcome of the first plurality of outcomes;

determine, based on the received first biomechanical data, a first plurality of sets of action-parameter values for respective action parameters of a plurality of action parameters, each of the first plurality of sets of action-parameter values corresponding to an action of the first plurality of actions, the first plurality of sets of action-parameter values being associated with respective outcomes of the first plurality of outcomes;

rank the plurality of action parameters for the first user, wherein rankings are determined based on a first analysis of the first plurality of sets of action-parameter values and of the respective outcomes of the first plurality of outcomes;

assign weights to the plurality of action parameters based on the rankings; and

generate a first athletic-performance model of the first user based on the first plurality of sets of action-parameter values, on the assigned weights of the plurality of action parameters, and on the first outcome data, wherein the first athletic-performance model comprises target ranges of action-parameter values computed for the first user based on a second analysis of the first plurality of sets of action-parameter values, of the assigned weights of the plurality of action parameters, and of the first outcome data.

19. The system of claim 18, wherein the one or more processors are further operable when executing the instructions to:

receive, from the one or more sensors in real time, real-time biomechanical data of the first user performing a real-time action of the first action-type and real-time outcome data of an outcome corresponding to the real-time action;

determine, based on the received real-time biomechanical data, an updated set of action-parameter values; and

dynamically update the first athletic-performance model based on the updated set of action-parameter values.