Apparatus and Method for Recognizing Spatial Gesture

• US 20150138088A1
• Filed: 11/18/2013
• Published: 05/21/2015
• Est. Priority Date: 09/09/2013
• Status: Active Grant

• Current Assignee: Korea Advanced Institute of Science and Technology
• Original Assignee: Center of Human-centered Interaction for Coexistence (Government of South Korea)
• Inventors: You, Bumjae, Yeom, Kiwon, Han, Hyejin
• Granted Patent: US 9,524,031 B2
• Time in Patent Office: Days
• US Class Current: 345/158
• 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: G06V 40/28 Recognition of hand or arm movements, e.g. recognition of deaf sign language static hand signs G06V40/113

Abstract

The present invention relates to an apparatus for recognizing a gesture in a space. In accordance with an embodiment, a spatial gesture recognition apparatus includes a pattern formation unit for radiating light onto a surface of an object required to input a gesture in a virtual air bounce, and forming a predetermined pattern on the surface of the object, an image acquisition unit for acquiring a motion image of the object, and a processing unit for recognizing a gesture input by the object based on the pattern formed on the surface of the object using the acquired image. In this way, depending on the depths of an object required to input a gesture in a space, haptic feedbacks having different intensities are provided, and thus a user can precisely input his or her desired gesture.

Claims

1. An apparatus for recognizing a spatial gesture, comprising:

a pattern formation unit for radiating light onto a surface of an object required to input a gesture in a virtual air bounce, and forming a predetermined pattern on the surface of the object;

an image acquisition unit for acquiring a motion image of the object; and

a processing unit for recognizing a gesture input by the object based on the pattern formed on the surface of the object using the acquired image.

2. The apparatus of claim 1, wherein the air bounce comprises:

a first area in which a plurality of virtual electrodes are arranged at regular intervals on a plane in a space, and in which a motion of the object is recognized as a gesture;

a second area in which an electric field is formed in a forward direction by the plurality of virtual electrodes; and

a third area in which the air bounce starts.

3. The apparatus of claim 2, further comprising a haptic provision unit for, if the object enters the air bounce to input a gesture, providing haptic feedbacks having one or more of different intensities and different types to the object in the first area, the second area, and the third area.

4. The apparatus of claim 1, wherein the processing unit comprises:

a pattern analysis unit for analyzing the pattern formed on the object and calculating depth information of the object entering the air bounce; and

a gesture recognition unit for recognizing a motion of the object made in a predetermined area of the air bounce as a gesture based on the calculated depth information of the object.

5. The apparatus of claim 4, wherein the pattern analysis unit calculates one or more of size information and density information of the pattern formed on the object, and calculates the depth information of the object using the calculated information.

6. The apparatus of claim 4, wherein the gesture recognition unit extracts information about one or more feature vectors from the recognized gesture, and eliminates a ligature from the motion of the object using the extracted feature vectors.

7. The apparatus of claim 6, wherein the gesture recognition unit extracts one or more segmentation points using the extracted feature vectors and the calculated depth information of the object, and determines a portion for connecting the extracted segmentation points to be a ligature.

8. The apparatus of claim 7, wherein the gesture recognition unit extracts the segmentation points by applying the feature vectors and the depth information of the object to any one of feature point extraction techniques including a matching probability technique and a pattern model technique.

9. A method of recognizing a spatial gesture, comprising:

radiating light onto a surface of an object required to input a gesture in a virtual air bounce, and forming a predetermined pattern on the surface of the object;

acquiring a motion image of the object; and

recognizing a gesture input by the object based on the pattern formed on the surface of the object using the acquired image.

10. The method of claim 9, wherein recognizing the gesture comprises:

analyzing the pattern formed on the object and calculating depth information of the object entering the air bounce,

wherein a motion of the object made in a predetermined area of the air bounce is recognized as a gesture based on the calculated depth information of the object.

11. The method of claim 10, wherein calculating the depth information is configured to calculate one or more of size information and density information of the pattern formed on the object, and calculate the depth information of the object using the calculated information.

12. The method of claim 10, wherein recognizing the gesture comprises:

extracting information about one or more feature vectors from the motion of the object; and

eliminating a ligature from the motion of the object using the extracted feature vectors.

13. The method of claim 12, wherein eliminating the ligature comprises:

extracting one or more segmentation points using the extracted feature vectors and the calculated depth information of the object; and

determining a portion for connecting the extracted segmentation points to be the ligature.

14. The method of claim 13, wherein extracting the segmentation points is configured to extract the segmentation points by applying the feature vectors and the depth information of the object to any one of feature point extraction techniques including a matching probability technique and a pattern model technique.