Hand gesture recognition

• US 8,768,006 B2
• Filed: 12/23/2010
• Issued: 07/01/2014
• Est. Priority Date: 10/19/2010
• Status: Active Grant

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Subramanian, Anbumani, Pathangay, Vinod, Mandalapu, Dinesh
• Primary Examiner(s): Seth, Manav
• Application Number:
  US12/977,065
• Publication Number:
  US 20120093360A1
• Time in Patent Office: 1,286 Days
• US Class Current: 382/103
• CPC Class Codes: G06V 10/143 Sensing or illuminating at different wavelengths
• CPC Class Codes: G06V 20/64 Three-dimensional objects
• 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

Systems, methods, and machine readable and executable instructions are provided for hand gesture recognition. A method for hand gesture recognition can include detecting, with an image input device in communication with a computing device, movement of an object. A hand pose associated with the moving object is recognized and a response corresponding to the hand pose is initiated.

Claims

1. A method for hand-gesture recognition, comprising:

detecting, with an image input device in communication with a computing device, movement of an object;

determining whether the object is a hand by directly employing a hue model learned from a plurality of images of generic skin color;

recognizing a hand pose associated with the moving object; and

initiating a response corresponding to the hand pose.

2. The method of claim 1, further comprising:

determining a color of the moving object; and

enabling recognition of the hand pose associated with the moving object if the color is determined to be substantially within a range of skin colors,

wherein recognizing the hand gesture includes recognizing the hand pose and a trajectory of the hand.

3. The method of claim 1, further comprising enabling recognition of a hand pose associated with the moving object if the color is determined to be substantially within a range of skin colors, wherein recognizing the hand gesture includes recognizing the hand pose and determining a trajectory of the hand pose.

4. The method of claim 1, wherein pixels from the image input device associated with the object having range information outside of a defined interaction volume with respect to the image input device are ignored.

5. The method of claim 1, further comprising:

initiating one mode of operation for the hand pose being classified as corresponding to a pointing hand pose; and

initiating another mode of operation for the hand pose not being classified as corresponding to a pointing hand pose.

6. The method of claim 5, wherein the another mode of operation includes:

initiating a fist response corresponding to for the hand pose being classified as corresponding to a fist gesture;

initiating a hand shake response corresponding to for the hand pose being classified as corresponding to a hand shake gesture;

initiating an open-palm response corresponding to for the hand pose being classified as corresponding to an open-palm gesture;

initiating a digit response corresponding to for the hand pose being classified as corresponding to a digit gesture;

initiating a thumbs up response corresponding to for the hand pose being classified as corresponding to a thumbs up gesture;

initiating a thumbs down response corresponding to for the hand pose being classified as corresponding to a thumbs down gesture; and

initiating an other response corresponding to for the hand pose not being classified as corresponding to any of the aforementioned gestures.

7. The method of claim 5, wherein the one mode of operation includes moving a cursor displayed on a production device in a direction and distance corresponding to movement of the pointing hand pose.

8. The method of claim 1, wherein detecting movement of the object is based on an optical flow analysis that includes initial thresholding based on disparity values associated with each pixel of the image input device.

9. The method of claim 7, further comprising:

after initial thresholding, segmenting the object detected as moving from a background scene, and

evaluating a color of the object including measuring a quantity of pixels producing a color in the range of skin colors and comparing the quantity of pixels to a threshold quantity.

10. The method of claim 8, wherein segmenting the object includes dynamic use of depth map thresholding in filtering pixels of the image input device outside a given depth range associated with an observed closest peak concentration of pixels.

11. The method of claim 1, wherein the first mode of operation includes moving a cursor in relation to the detected movement of the object, the cursor being displayed on a production device communicatively coupled to the computing device.

12. A hand gesture recognition system, comprising:

a computing device having at least one processor;

an image input device communicatively coupled to the computing device, the image input device providing per pixel color information and pixel range information;

a production device communicatively coupled to the computing device; and

a non-transitory computer-readable medium having computer-readable instructions stored thereon that, if executed by the at least one processor, cause the at least one processor to;

detect a moving object within an interaction zone of the image input device;

evaluate color of the moving object from the color information;

enable hand gesture recognition of the moving object evaluated as exceeding a threshold portion of a particular color, the hand gesture recognition to continuously attempt to recognize a hand gesture associated with the moving object as any of a plurality of hand gestures associated with different operations, directly responsive to the moving object exceeding the threshold portion; and

wherein hand gesture recognition includes hand pose and moving object trajectory determinations, and wherein the interaction zone is defined at least in terms of the range information.

13. The system of claim 12, wherein the image input device is a depth camera providing a color image and a disparity image.

14. The system of claim 12, wherein the computer-readable instructions further cause the at least one processor to implement a probabilistic support vector machine utilized to classify hand gestures.

15. A non-transitory computer-readable medium having computer-readable instructions stored thereon that, if executed by one or more processors, cause the one or more processors to:

detect, with an image input device in communication with a computing device, movement of an object;

evaluate a color of the object with respect to a range of skin colors in direct conjunction with a hue model learned from a plurality of images of generic skin colors; and

enable recognition of a hand gesture if the color of the object is within the range of skin colors the recognition to continuously attempt to recognize the hand gesture associated with the object as any of a plurality of hand gestures associated with different operations, directly responsive to the color of the object being within the range of skin colors.