Systems and metohds of gesture recognition

Abstract

Disclosed methods include a method of controlling a computing device includes the steps of detecting a gesture made by a human user, identifying the gesture, and executing a computer command. The gesture may comprise a change in depth of a body part of the human user relative to the 2D camera. The gesture may be detected via a 2D camera in electronic communication with the computing device. Disclosed systems include a 2D camera and a computing device in electronic communication therewith. The 2D camera is configured to capture at least a first and second image of a body part of a human user. The computing device is configured to recognize at least a first object in the first image and a second object in the second image, identify a change in depth, and execute a command in response to the change in depth.

What is claimed is: 
     
         1 . A method of controlling a computing device, the method comprising:
 detecting, via a 2D camera in electronic communication with the computing device, a gesture made by a human user of the computing device;   wherein the gesture comprises a change in depth of a body part of the human user relative to the 2D camera wherein depth is a dimension that is measured parallel to a line extending perpendicularly from a lens of the 2D camera; and   wherein the detecting step comprises capturing at least a first image of the body part in an initial gesture position at an initial depth and a second image of the body part in a subsequent gesture position at a subsequent depth, wherein the subsequent depth is different than the initial depth;   identifying the gesture via a computing unit of the computing device, wherein the identifying step comprises:
 recognizing at least a first object in the body part of the first image and a second object in the body part of the second image, wherein the recognizing step comprises; 
 generating a first boundary in the first image that fully contains the first object, wherein the first object corresponds to the body part in the initial gesture position; 
 generating a second boundary in the second image that fully contains the second object, wherein the second object corresponds to the body part in the subsequent gesture position; 
 determining the change in a magnitude depth from the initial gesture position to the subsequent gesture position without requiring determination of an initial magnitude depth or a subsequent magnitude depth, wherein the determining comprises comparing a size of the first boundary to a size of the second boundary; 
   converting the gesture into a command via the computing unit; and   executing, in response to the converting step, the command via the computing unit.   
     
     
         2 . The method of  claim 1 , wherein the recognizing step includes using an object detection algorithm to recognize the first and second images. 
     
     
         3 . The method of  claim 1 , wherein the first boundary is the smallest region of the first image that fully contains the body part of the human user, and wherein the second boundary is the smallest region of the second image that fully contains the body part of the human user. 
     
     
         4 . The method of  claim 3 , wherein the object detection algorithm is selected from the group consisting of a Haar Cascade Classifier and a HOG Detector Classifier. 
     
     
         5 . The method of  claim 1 , wherein:
 when the first object is smaller than the second object, the gesture is identified as a forward moving gesture; and   when the first object is larger than the second object, the gesture is identified as a rearward moving gesture.   
     
     
         6 . The method of  claim 5 , comprising converting a rearward moving gesture into a command to decrease a size of an image on a display unit in electronic communication with the computing device. 
     
     
         7 . The method of  claim 5 , comprising converting a foreword moving gesture into a command to increase a size of an image on a display unit in electronic communication with the computing device. 
     
     
         8 . The method of  claim 1 , wherein the comparing step includes quantifying a difference in at least one of the width of the first boundary and the second boundary and the height of the first boundary and the second boundary using a scaling constant; and 
     
     
         9 . The method of  claim 8 , wherein the scaling constant is changed dynamically by dividing the change in magnitude depth by the time difference between the first boundary and second boundary. 
     
     
         10 . The method of  claim 8 , wherein the converting step includes generating a magnitude associated with the command, wherein the magnitude is based on the quantified difference in the at least one of the width of the first boundary and the second boundary and the height of the first boundary and the second boundary 
     
     
         11 . The method of  claim 10 , wherein the comparing step includes analyzing a speed of the gesture, and wherein the converting step includes adjusting the magnitude associated with the command based on the speed of the gesture. 
     
     
         12 . The method of  claim 1 , wherein the initial gesture position comprises a particular orientation of the body part of the human user. 
     
     
         13 . The method of  claim 12 , wherein the initial gesture position is one of an open hand and a closed hand. 
     
     
         14 . A system comprising:
 a 2D camera configured to capture at least a first and second image of a gesture made by a human user, wherein the gesture comprises a change in depth of a body part of the user relative to the 2D camera, and wherein depth is a dimension that is measured parallel to a line extending perpendicularly from a lens of the 2D camera;   a computing device in electronic communication with the 2D camera configured to:
 recognize at least a first object in the body part of the first image and a second object in the body part of the second image, wherein the first object corresponds to the body part in an initial gesture position at an initial depth and the second object corresponds to the body part in a subsequent gesture position at a subsequent depth, and wherein the subsequent depth is different than the initial depth; 
 generate a first boundary in the first image, the first boundary being the a smallest region of the first image that fully contains the first object; 
 generate a second boundary in the second image, the second boundary being the a smallest region of the second image that fully contains the second object; 
 compare at least a size of the first boundary to a size of the second boundary in order to determine a magnitude of the change in depth of the body part of the human user relative to the 2D camera without requiring determination of an absolute magnitude of the initial depth or an absolute magnitude of the subsequent depth; 
 convert the change in depth of the body part into a command; and 
 execute the command. 
   
     
     
         15 . The system of  claim 14 , wherein the computing device is configured to recognize the first and second objects via an object detection algorithm. 
     
     
         16 . The system of  claim 14 , wherein the computing device is configured to identify a first gesture in a first initial gesture position. 
     
     
         17 . The system of  claim 16 , wherein the computing device is configured to identify a second gesture in a second initial gesture position and terminate the first command in response to the identification of the second gesture. 
     
     
         18 . The system of  claim 14 , wherein the 2D camera is integral to the computing device. 
     
     
         19 . The system of  claim 14 , wherein the 2D camera is remote of the computing device. 
     
     
         20 . The system of  claim 14 , comprising a display unit in electronic communication with the computing device configured to display a virtual image to the human user.