Systems and methods for performing anytime motion recognition

• US 8,156,067 B1
• Filed: 04/29/2010
• Issued: 04/10/2012
• Est. Priority Date: 07/14/2006
• Status: Active Grant

• Current Assignee: AiLive, Inc.
• Original Assignee: AiLive, Inc.
• Inventors: Powers, III, William Robert, Reynolds, Stuart, Wilkinson, Dana, Wright, Ian, Yen, Wei, Tu, Xiaoyuan, KAWANO, Yoichiro, Musick, JR., Charles
• Primary Examiner(s): Chaki, Kakali
• Assistant Examiner(s): Rifkin, Ben
• Application Number:
  US12/770,350
• Time in Patent Office: 712 Days
• Field of Search: 706/20, 706/47
• US Class Current: 706/47
• CPC Class Codes: A63F 13/211 using inertial sensors, e.g. accelerometers or gyroscopes
• CPC Class Codes: A63F 13/2145 the surface being also a display device, e.g. touch screens
• CPC Class Codes: A63F 13/215 comprising means for detecting acoustic signals, e.g. using a microphone
• CPC Class Codes: A63F 13/235 using a wireless connection, e.g. infrared or piconet
• CPC Class Codes: A63F 13/428 involving motion or position input signals, e.g. signals representing the rotation of an input controller or a player's arm motions sensed by accelerometers or gyroscopes
• CPC Class Codes: A63F 13/67 adaptively or by learning from player actions, e.g. skill level adjustment or by storing successful combat sequences for re-use
• CPC Class Codes: A63F 13/92 Video game devices specially adapted to be hand-held while playing
• CPC Class Codes: A63F 2300/1043 being characterized by constructional details
• CPC Class Codes: A63F 2300/105 using inertial sensors, e.g. accelerometers, gyroscopes
• CPC Class Codes: A63F 2300/6027 using adaptive systems learning from user actions, e.g. for skill level adjustment
• CPC Class Codes: A63F 2300/6045 for mapping control signals received from the input arrangement into game commands
• CPC Class Codes: G06F 2203/0384 Wireless input, i.e. hardware and software details of wireless interface arrangements for pointing devices
• 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
• CPC Class Codes: G06F 3/0346 with detection of the device orientation or free movement in a 3D space, e.g. 3D mice, 6-DOF [six degrees of freedom] pointers using gyroscopes, accelerometers or tilt-sensors
• CPC Class Codes: G06F 3/038 Control and interface arrangements therefor, e.g. drivers or device-embedded control circuitry

Abstract

Techniques for performing anytime motion recognition are disclosed. Based on a motion signal received from a motion-sensitive device being manipulated by a user as a way to interact with an application, a processing unit is configured to a set of prototypes included in a motion recognizer to generate a recognition signal from the motion signal in response to the motion recognizer. A prototype is eliminated out of the set of the prototypes from further consideration when a corresponding running distance of the prototype to the motion signal is beyond a predefined measurement so that the motion recognition speeds up as time passes.

Claims

1. A method for performing anytime motion recognition, the method comprising:

receiving a motion signal about a motion from a motion-sensitive device being manipulated by a user as a way to interact with an application;

generating a final recognition signal for the motion signal in response to a motion recognizer which includes a set of prototypes, wherein said generating a final recognition signal comprises identifying a current node in a disambiguation tree to represent a set of moves currently confused with each other given how much of the motion signal has been received so far, wherein the disambiguation tree is prebuilt before receiving the motion signal; and

returning the current node to the application at any point in time before the motion signal is received in entirety so that the current node facilitates the application to start an immediate animation of the motion, even when moves are confused.

2. The method as recited in claim 1, wherein the disambiguation tree is used as part of a feedback process attached to the motion recognizer.

3. The method as recited in claim 1, wherein the disambiguation tree is attached to either prebuilt motion recognizers or ad-hoc recognizers and allow to work with the user in proxy to help ensure the user is building the ad-hoc motion recognizers that fit an early animation decision of the application.

4. The method as recited in claim 3, wherein the disambiguation tree tells the application an earliest point in time when it is safe to begin animating for a specific set of moves, and when it is safe to commit to a single move.

5. The method as recited in claim 1, further comprising:

generating the final recognition signal anytime even if only a first portion of data has been considered when the current node indicates it is safe to commit to a single move.

6. A system for performing anytime motion recognition, the system comprising:

at least one motion-sensitive device;

a processing unit receiving a motion signal from the motion-sensitive device when the motion-sensitive device is being manipulated by a user, the processing unit being loaded with at least one motion recognizer including a set of prototypes, and configured to;

generate a recognition signal from the motion signal in response to the motion recognizer,

identify a current node in a disambiguation tree to represent a set of moves currently confused with each other given how much of the motion signal has been received so far, wherein the disambiguation tree is prebuilt before receiving the motion signal; and

returning a current node to the application at any point in time before the motion signal is received in entirety so that the current node facilitates the application to start an immediate animation of the motion, even when moves are confused.

7. The system as recited in claim 6, wherein the disambiguation tree is used as part of a feedback process attached to the motion recognizer.

8. The system as recited in claim 6, wherein the disambiguation tree is attached to either prebuilt motion recognizers or ad-hoc recognizers and allow to work with the user in proxy to help ensure the user is building the ad-hoc motion recognizers that fit an early animation decision of the application.

9. The system as recited in claim 8, the disambiguation tree tells the application an earliest point in time when it is safe to begin animating for a specific set of moves, and when it is safe to commit to a single move.

10. The system as recited in claim 6, wherein the processing unit is further configured to generate the final recognition signal anytime even if only a first portion of data has been considered when the current node indicates it is safe to commit to a single move.

11. A method for performing anytime motion recognition, the method comprises:

receiving a motion signal from a motion-sensitive device being manipulated by a user as a way to interact with an application;

accessing at least a motion recognizer embedded with a disambiguation tree that organizes a collection of sets of moves that are commonly confused with each other at certain points in time;

identifying a current node in the disambiguation tree that represents a set of moves that are currently confused with each other given how much of the motion signal has been received so far; and

returning a set of currently confused moves to the application at any point in time to start an immediate animation of a move before a final recognition signal to the motion signal is determined.

12. The method as recited in claim 11, further comprising starting to animate the move using either preloaded motion recognizers or ad-hoc motion recognizers, even when the move is confused with other moves.

13. The method as recited in claim 12, wherein the move facilitates an end user to build the ad-hoc motion recognizers that fit early animation decisions of the application.

14. The method as recited in claim 12, wherein the move provides an instant response to a user as to which moves need to be redesigned in response to information of the disambiguation tree'"'"'s on exactly which moves stay confused for how long.

15. A system for performing anytime motion recognition, the system comprising:

at least one motion-sensitive device;

a processing unit receiving a motion signal from the motion-sensitive device when the motion-sensitive device is being manipulated by a user interacting with an application, the processing unit being loaded with at least one motion recognizer embedded with a disambiguation tree that organizes a collection of sets of moves

that are commonly confused with each other at certain points in time, and wherein the processing unit is configured to identify a current node in the

disambiguation tree that represents a set of moves that are currently confused with each other; and

return the set of currently confused moves to the application at any point in time to start an immediate animation of a move before a final recognition signal to the motion signal is determined.

16. The system as recited in claim 15, wherein the processing unit is configured to animate a move using either preloaded motion recognizers or ad-hoc motion recognizers, even when the move is confused with other moves.

17. The system as recited in claim 16 wherein the move facilitates an end user to build the ad-hoc motion recognizers that fit early animation decisions of the application.

18. The system as recited in claim 17, wherein the move provides an instant response to a user as to which moves need to be redesigned in response to information of the disambiguation tree on exactly which moves stay confused for how long.