System and method for object recognition and tracking in a video stream

Abstract

The invention provides a system method for object detection and tracking in a video stream. Frames of the video stream are divided into regions of interest and a probability is calculated for each region of interest that the region contains at least a portion of an object to be tracked. The regions of interest in each frame are then classified based on the calculated probabilities. A region of interest (RI) frame is then constructed for each video frame that reports the classification of regions of interest in the video frame. Two or more RI frames are then compared in order to determine a motion of the object. The invention also provides a system executing the method of the invention, as well as a device comprising the system. The device may be for example, a portable computer, a mobile telephone, or an entertainment device.

1 .- 30 . (canceled) 
     
     
         31 . A system comprising:
 a processor configured to:
 partition a first video frame in a video stream into at least a first region; 
 classify the first region based on an analysis of one or more pixels within the first region, wherein a classification of the first region reflects a probability of a presence of a tracked object within the first region; 
 compare the classification of the first region with a classification of another region of another video frame in the video stream; and 
 based on the comparison of the classification of the first region and the classification of another region of another video frame in the video stream, determine a motion of the tracked object. 
   
     
     
         32 . The system of  claim 31 , wherein to partition the first video frame the processor is further configured to partition the first video frame based on an axis of expected motion of the tracked object. 
     
     
         33 . The system of  claim 31 , wherein to partition the first video frame the processor is further configured to:
 analyze one or more video frames in the video stream;   determine, based on an analysis of the one or more video frames, one or more high variance regions within the one or more video frames; and   omit the one or more high variance regions from the first region.   
     
     
         34 . The system of  claim 31 , wherein to classify the first region the processor is further configured to classify the first region based on an analysis of one or more regions of one or more other video frames that precede the first video frame in the video stream. 
     
     
         35 . The system of  claim 31 , wherein to classify the first region the processor is further configured to classify the first region based on a distance between a histogram of the tracked object and a histogram of the first region. 
     
     
         36 . The system of  claim 31 , wherein to determine a motion of the tracked object the processor is further configured to detect a motion pattern of the tracked object based on the classification of the first region and one or more classifications of one or more other regions of one or more other video frames in the video stream. 
     
     
         37 . The system of  claim 31 , wherein to determine a motion of the tracked object the processor is further configured to:
 apply a pattern recognition test to the first region and one or more other regions of in the video stream; and   compute a probability that a motion pattern associated with the pattern recognition test occurred during a time window that includes the first video frame and the one or more other video frames.   
     
     
         38 . The system of  claim 31 , wherein to determine a motion of the tracked object the processor is further configured to detect a motion pattern of the tracked object based on the classification of the first region and one or more inputs from an operating system. 
     
     
         39 . The system of  claim 31 , wherein to determine a motion of the tracked object the processor is further configured to detect a motion pattern of the tracked object based on the classification of the first region and one or more inputs from an application. 
     
     
         40 . The system of  claim 31 , wherein the processor is further configured to execute a command associated with the determined motion of the tracked object. 
     
     
         41 . A method comprising:
 partitioning a first video frame in a video stream into at least a first region;   classifying the first region based on an analysis of one or more pixels within the first region, wherein a classification of the first region reflects a probability of a presence of a tracked object within the first region;   comparing the classification of the first region with a classification of another region of another video frame in the video stream; and   based on the comparison of the classification of the first region and the classification of another region of another video frame in the video stream, determining, by a processor, a motion of the tracked object.   
     
     
         42 . The method of  claim 41 , wherein partitioning the first video frame further comprises partitioning the first video frame based on an axis of expected motion of the tracked object. 
     
     
         43 . The method of  claim 41 , wherein partitioning the first video frame further comprises:
 analyzing one or more video frames in the video stream;   determining, based on an analysis of the one or more video frames, one or more high variance regions within the one or more video frames; and   omitting the one or more high variance regions from the first region.   
     
     
         44 . The method of  claim 41 , wherein classifying the first region further comprises classifying the first region based on an analysis of one or more regions of one or more other video frames that precede the first video frame in the video stream. 
     
     
         45 . The method of  claim 41 , wherein classifying the first region further comprises classifying the first region based on a distance between a histogram of the tracked object and a histogram of the first region. 
     
     
         46 . The method of  claim 41 , wherein determining a motion of the tracked object further comprises detecting a motion pattern of the tracked object based on the classification of the first region and one or more classifications of one or more other regions of one or more other video frames in the video stream. 
     
     
         47 . The method of  claim 41 , wherein determining a motion of the tracked object further comprises:
 applying a pattern recognition test to the first region and one or more other regions of in the video stream; and   computing a probability that a motion pattern associated with the pattern recognition test occurred during a time window that includes the first video frame and the one or more other video frames.   
     
     
         48 . The method of  claim 41 , wherein determining a motion of the tracked object further comprises detecting a motion pattern of the tracked object based on the classification of the first region and one or more inputs from an operating system. 
     
     
         49 . The method of  claim 41 , wherein determining a motion of the tracked object further comprises detecting a motion pattern of the tracked object based on the classification of the first region and one or more inputs from an application. 
     
     
         50 . A non-transitory computer readable medium having instructions encoded thereon that, when executed by a processing device, cause the processing device to:
 partition a first video frame in a video stream into at least a first region;   classify the first region based on an analysis of one or more pixels within the first region, wherein a classification of the first region reflects a probability of a presence of a tracked object within the first region;   compare the classification of the first region with a classification of another region of another video frame in the video stream; and   based on the comparison of the classification of the first region and the classification of another region of another video frame in the video stream, determine, by the processing device, a motion of the tracked object.