Gesture input system, method and program

• US 9,032,336 B2
• Filed: 09/07/2006
• Issued: 05/12/2015
• Est. Priority Date: 09/07/2006
• Status: Active Grant

• Current Assignee: Osaka Electro-Communication University
• Original Assignee: Osaka Electro-Communication University
• Inventors: Kimura, Katsutoshi, Uoi, Hirotaka
• Primary Examiner(s): Pillai, Namitha
• Application Number:
  US12/440,150
• Publication Number:
  US 20100138785A1
• Time in Patent Office: 3,169 Days
• Field of Search: 715/834, 715/863, 715/810, 715/849, 715/852, 345/655
• US Class Current: 715/863
• 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

The present invention provides an input interface that a human can operate naturally and intuitively using a human gesture (motion), by including means for acquiring motion information based on a gesture and input interface means for generating information for operating an object on a desktop of a computer on the basis of motion information. In this case, the motion information is matched against a template for recognizing a motion of a user and a matched event is outputted so that the object is operated. The object includes a pie menu in which a menu item is disposed in a circular form. A user is allowed to selects a desired menu item in the pie menu in accordance with an angle at which the user twists a wrist thereof.

Claims

1. A gesture input system comprising:

motion information acquisition means for acquiring motion information of an individual based on a gesture of the individual;

preprocessing means for performing normalization and position-coordinate conversion on motion information acquired by the motion information acquisition means, wherein the normalization is performed based at least in part of a ratio of body data of the individual that made the gesture to body data of a predetermined human body model;

template matching means for matching, based on one or more parameters, the normalized and converted motion information against a start template and an end template to recognize a start and an end of a motion automatically, wherein if the normalized and converted motion information is matched to the start template, a motion start event is generated, further wherein the one or more parameters comprise position, velocity, acceleration, distance, and direction; and

input interface means for generating information for operating an object on a desktop of a computer on the basis of the normalized and converted motion information as matched by the template matching means.

2. The gesture input system according to claim 1, further comprising

template storage means for storing the start template and the end template for recognizing the motion of the individual;

and

event output means for outputting an output event on the basis of the motion information as matched by the template matching means.

3. The gesture input system according to claim 2, wherein the object includes a pie menu in which a menu item is disposed in a circular form, and a user selects any menu item in the pie menu by twisting a wrist thereof.

4. The gesture input system according to claim 3, wherein the event output means sequentially selects a menu item in the pie menu in a direction in which a user twists a wrist thereof.

5. The gesture input system according to claim 4, wherein the pie menu is layered.

6. The gesture input system of claim 1, wherein the motion information is acquired at least in part from one or more markers mounted to the individual.

7. The gesture input system of claim 6, wherein the one or more markers are configured to reflect light.

8. The gesture input system of claim 1, wherein the object on the desktop comprises a first cursor and a second cursor.

9. The gesture input system of claim 8, wherein the first cursor corresponds to a right hand of the individual and the second cursor corresponds to a left hand of the individual.

10. The gesture input system of claim 8, wherein the first cursor and the second cursor each have a normal state and a selected state, and wherein the selected state is activated by the gesture of the individual.

11. The gesture input system of claim 1, wherein operating the object on the desktop of the computer increases or decreases a window size on the desktop.

12. The gesture input system of claim 1, wherein the position-coordinate conversion is performed based at least in part on the orientation of the individual.

13. The gesture input system of claim 1, wherein another one of the motion start event is not generated until the motion information is matched with the end template.

14. A gesture input method comprising:

acquiring motion information of an individual based on a gesture of the individual;

performing normalization and position-coordinate conversion on the acquired motion information, wherein the normalization is performed based at least in part on a ratio of body data of the individual that made the gesture to body data of a predetermined human body model;

matching, based on one or more parameters, the normalized and converted motion information against a start template and an end template to recognize a start and an end of a motion automatically, wherein if the normalized and converted motion information is matched to the start template, a motion start event is generated, further wherein the one or more parameters comprise position, velocity, acceleration, distance, and direction; and

generating information for operating an object on a desktop on the basis of the normalized and converted motion information that is matched to the start template and the end template.

15. The method of claim 14, wherein matching the acquired motion information to the start template and the end template is based on one or more parameters.

16. The method of claim 15, wherein the one or more parameters comprise position, velocity, acceleration, distance, and direction.

17. The method of claim 14, wherein one or more markers are mounted to the individual, and wherein at least a portion of the motion information is acquired based on received light reflections from the one or more markers.

18. The method of claim 14, wherein operating the object on the desktop of the computer comprises increasing or decreasing a window size on the desktop.

19. The method of claim 18, wherein increasing or decreasing the window size on the desktop is based on a one-handed gesture from the individual.

20. A non-transitory computer-readable medium having computer-readable instructions stored thereon, wherein the computer-readable instructions cause a computer to:

acquire motion information of an individual based on a gesture of the individual;

perform normalization and position-coordinate conversion on the acquired motion information, wherein the normalization is performed based at least in part on a ratio of body data of the individual that made the gesture to body data of a predetermined human body model;

match, based on one or more parameters, the normalized and converted motion information against a start template and an end template to recognize a start and an end of a motion automatically, wherein if the normalized and converted motion information is matched to the start template, a motion start event is generated, further wherein the one or more parameters comprise position, velocity, acceleration, distance, and direction; and

generate information for operating an object on a desktop on the basis of the normalized and converted motion information that is matched to the start template and the end template.

21. The non-transitory computer-readable medium of claim 20, wherein the object on the desktop comprises a first cursor and a second cursor, and wherein the first cursor corresponds to a right hand of the individual and the second cursor corresponds to a left hand of the individual.