Gesture based human interfaces

• US 8,837,780 B2
• Filed: 06/22/2012
• Issued: 09/16/2014
• Est. Priority Date: 06/22/2012
• Status: Active Grant

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Jia, Charles, Eades, Lynn M., Lasslo, Laurel, Arrioia, Christopher, Rose, Michael Stephen
• Primary Examiner(s): NAKHJAVAN, SHERVIN K
• Application Number:
  US13/531,270
• Publication Number:
  US 20130343601A1
• Time in Patent Office: 816 Days
• Field of Search: None
• US Class Current: 382/103
• 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: G06V 10/143 Sensing or illuminating at different wavelengths
• 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 method for implementing gesture based human interfaces includes segmenting data generated by an IR camera of an active area and detecting objects in an active area. The objects are distinguished as either island objects or peninsula objects and a human hand is identified from among the peninsula objects. The motion of the human hand is tracked as a function of time and a gesture made by the human hand is recognized.

Claims

1. A method for implementing a human based interface comprising:

segmenting, by a computing device, data generated by an IR camera of an active area;

detecting, by a computing device, objects in the active area;

distinguishing, by a computing device, between island objects and peninsula objects in the active area;

identifying, by a computing device, a human hand from among the peninsula objects, wherein identifying a peninsula object comprises determining that an object intersects a border of the active area;

tracking, by a computing device, motion of the human hand in the active area as a function of time; and

recognizing, by a computing device, a gesture made by the human hand.

2. The method of claim 1, in which segmenting data comprises edge detection.

3. The method of claim 1, in which segmenting data further comprises segmenting data generated by a depth camera.

4. The method of claim 3, in which segmenting data comprises combining data from the IR camera, a visible camera and depth camera and operating on the combined data.

5. The method of claim 1, in which segmenting data comprises distinguishing light reflected from real objects and visible light projected by a projector onto the active area using data generated by the IR camera.

6. The method of claim 1, in which detecting objects in the active area comprises comparing edges detected from IR camera with height contours detected by the depth camera.

7. The method of claim 1, in which identifying an island object comprises determining that object has an edge that forms a complete perimeter within the active area.

8. The method of claim 1, in which detecting objects in the active area comprises:

illuminating the active area with at a frequency of IR light; and

distinguishing human skin based on reflectance of the frequency of IR light.

9. The method of claim 8, in which distinguishing human skin comprises comparison of two IR wavelengths, a first wavelength in which human skin has a higher IR reflectivity and a second wavelength in which human skin has a lower IR reflectivity.

10. The method of claim 1, further comprising:

interpreting the gesture as a event;

passing the event to an application; and

execution of a command within the application as a function of the event.

11. A gesture based human interface system comprising:

a projector to project an optical image over an active area;

a color camera imaging objects in the active area and the optical image projected over the active area;

infrared (IR) illumination;

an IR camera sensitive to the IR illumination reflected off objects in the active area but not to the optical image projected over the active area;

a computer processor to receive input from the IR and color cameras and process the data to distinguish between island objects and peninsula objects in the active area, to identify a human hand from among the peninsula objects, wherein the identifying of the peninsula object comprises determining that an object edge intersects a border of the active content, and to detect gestures by the human hand.

12. The system of claim 11, in which the IR illumination comprises a narrow waveband illumination in which human skin has a high reflectance and the IR camera comprises a filter to selectively pass the narrow waveband illumination reflected from human skin.

13. A computer program product for implementing a human based interface comprising:

a non-transitory computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising;

computer readable program code to segment data generated by an IR camera of an active area;

computer readable program code to detect objects in the active area;

computer readable program code to distinguish between island objects and peninsula objects in the active area; and

computer readable program code to identify a human hand from among the peninsula objects, wherein identifying a peninsula object comprises determining that an object edge intersects a border of the active area.