ISOLATE EXTRANEOUS MOTIONS

• US 20100278393A1
• Filed: 05/29/2009
• Published: 11/04/2010
• Est. Priority Date: 05/01/2009
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Geisner, Kevin, Markovic, Relja, Snook, Gregory Nelson, Latta, Stephen Gilchrist
• Granted Patent: US 8,942,428 B2
• Time in Patent Office: Days
• US Class Current: 382/107
• CPC Class Codes: A63F 13/213 comprising photodetecting means, e.g. cameras, photodiodes or infrared cells A63F13/219, A63F13/655 take precedence
• CPC Class Codes: G06F 2203/012 Walk-in-place systems for allowing a user to walk in a virtual environment while constraining him to a given position in the physical environment
• CPC Class Codes: G06F 3/011 Arrangements for interaction with the human body, e.g. for user immersion in virtual reality blind teaching G09B21/00
• 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: G06V 40/20 Movements or behaviour, e.g. gesture recognition recognition of facial expressions G06V40/16
• 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

A system may receive image data and capture motion with respect to a target in a physical space and recognize a gesture from the captured motion. It may be desirable to isolate aspects of captured motion to differentiate random and extraneous motions. For example, a gesture may comprise motion of a user'"'"'s right arm, and it may be desirable to isolate the motion of the user'"'"'s right arm and exclude an interpretation of any other motion. Thus, the isolated aspect may be the focus of the received data for gesture recognition. Alternately, the isolated aspects may be an aspect of the captured motion that is removed from consideration when identifying a gesture from the captured motion. For example, gesture filters may be modified to correspond to the user'"'"'s natural lean to eliminate the effect the lean has on the registry of a motion with a gesture filter.

Claims

1. A method comprising:

receiving data of a scene, wherein the data includes data representative of a user'"'"'s captured motion;

collecting a history of the data representative of the user'"'"'s captured motion;

identifying a recurring feature in the history of data;

analyzing the data with base information representing a gesture to produce an output corresponding to whether the data corresponds to the gesture;

adjusting the base information representing the gesture to correspond to the recurring feature in the history of data representative of the user'"'"'s captured motion.

2. The method of claim 1, wherein adjusting the base information comprises setting at least one parameter of the filter to correspond to the recurring feature.

3. The method of claim 1, further comprising predicting an intended gesture from the received data, wherein the filter representing a gesture is a filter representing the intended gesture.

4. The method of claim 1, further comprising applying the adjusted filter to the received data and identifying the gesture corresponding to the adjusted filter.

5. The method of claim 1, wherein the data representative of the user'"'"'s captured motion, that did not register with the filter prior to adjusting the base information, registers with the filter after adjusting the base information in the filter.

6. The method of claim 1, wherein the recurring feature identified from the history of the data is a detectable characteristic of a user recurring in the received data.

7. The method of claim 6, wherein the detectable characteristic is at least one of a motion of a user'"'"'s body part, a user'"'"'s posture, or a user'"'"'s size.

8. The method of claim 7, wherein adjusting the base information to correspond to the recurring feature comprises setting at least one parameter of the filter to correspond to the recurring motion of the user'"'"'s body part, the user'"'"'s posture, or the user'"'"'s size.

9. The method of claim 1, further comprising adjusting at least one property of a virtual object to correspond to the recurring feature.

10. The method of claim 9, wherein the recurring feature is a user'"'"'s size and adjusting the at least one property of the virtual object to correspond to the recurring feature comprises scaling dimensions of the virtual object to correspond to the user'"'"'s size.

11. The method of claim 1, wherein the filter is a first filter and is complementary with at least one other filter, the method further comprising adjusting the base information of the at least one other filter determined from the adjustment to the first filter.

12. The method of claim 11, wherein, prior to adjusting the base information of the first filter, the received data registers with the first filter, and wherein after adjusting the base information of the first filter and the at least one other filter, the received data registers with the at least one other filter.

13. A method comprising:

receiving data captured by a camera, wherein the received data is representative of a user'"'"'s motion in a physical space;

predicting an intended gesture from the received data;

selecting a first portion of the received data that is applicable to the intended gesture and excluding a second portion of the received data that is not applicable to the intended gesture;

providing a filter representing the intended gesture, the filter comprising base information about the intended gesture; and

applying the filter to the selected first portion of the received data and determining an output from the base information about the intended gesture.

14. The method of claim 13, further comprising applying a plurality of filters to the received data, wherein the predicted intended gesture is a gesture corresponding to at least one of the plurality of filters having base information that corresponds to the received data.

15. The method of claim 13, further comprising generating a model of a user from the received data, wherein the model maps to the first portion of the received data that is applicable to the intended gesture and comprises a pre-authored animation that represents the second portion of the received data

16. The method of claim 13, wherein selecting the first portion of the received data that is applicable to the intended gesture comprises selecting a portion of the received data that corresponds to a volume of interest.

17. The method of claim 13, further comprising adjusting the filter representing the intended gesture to a recurring feature in a history of data representative of the user'"'"'s motion, wherein selecting the first portion of the received data comprises selecting a portion of the received data that corresponds to the adjusted filter.

18. The method of claim 13, wherein the second portion of the received data that is not applicable to the intended gesture corresponds to an extraneous motion in the physical space.

19. The method of claim 18, wherein the extraneous motion is a motion of at least one of a second user, an extraneous body part, or an object in the physical space.

20. The method of claim 18, wherein the extraneous motion is a motion in a volume of interest.

21. The method of claim 18, wherein the output is indicative of a failure of the received data to register with the filter corresponding to the intended gesture.

22. The method of claim 18, wherein the output is a recognition of the intended gesture from the selected first portion of the received data.

23. The method of claim 22, further comprising aborting gesture recognition of the intended gesture when the first portion of the received data deviates from the base information in the filter.