Methods of modifying search areas

Abstract

A video encoding method includes receiving a video captured by an image capture device on-board a movable object, where the video includes a video frame component; receiving sensor data from a plurality of sensors on-board the movable object; generating, according to the sensor data, an optical flow field associated with the video frame component; and evaluating motion of the video for video encoding based on the optical flow field.

What is claimed is: 
     
         1 . A video encoding method comprising:
 receiving a video captured by an image capture device on-board a movable object, the video including a video frame component;   receiving sensor data from a plurality of sensors on-board the movable object;   generating, according to the sensor data, an optical flow field associated with the video frame component; and   evaluating motion of the video for video encoding based on the optical flow field.   
     
     
         2 . The method of  claim 1 , wherein evaluating the motion of the video includes:
 identifying an object within the video frame component; and   assessing, based on the optical flow filed, a motion vector of the object.   
     
     
         3 . The method of  claim 2 , further comprising determining an initial point for assessing the motion vector within the video frame component. 
     
     
         4 . The method of  claim 3 , wherein:
 the video frame component is a first video frame component of the video; and   determining the initial point includes determining the initial point by referencing a second video frame component of the video adjacent to the first video frame component.   
     
     
         5 . The method of  claim 1 , wherein evaluating the motion of the video includes:
 determining, based on the optical flow field, a portion of the motion of the video that is caused by a movement of the image capture device.   
     
     
         6 . The method of  claim 1 , wherein the plurality of sensors include at least one of an image sensor, a GPS, an inertial measurement unit, a lidar, an ultrasonic sensor, an acoustic sensor, or a WiFi sensor. 
     
     
         7 . The method of  claim 1 , wherein the sensor data includes information about a spatial disposition of the image capture device, a movement of the image capture device, a zoom state of the image capture device, and a field of view (FOV) of the image capture device. 
     
     
         8 . The method of  claim 1 , wherein generating the optical flow field includes generating the optical flow filed further according to data received from a source external to the movable object. 
     
     
         9 . The method of  claim 8 , wherein the data received from the source external to the movable object include data transmitted from a ground station to the movable object. 
     
     
         10 . The method of  claim 1 , wherein evaluating the motion of the video includes evaluating the motion of the video within a search area in the video frame component of the video. 
     
     
         11 . The method of  claim 10 , further comprising:
 determining a dimension, shape, or location of the search area based on the optical flow field.   
     
     
         12 . The method of  claim 11 , further comprising:
 obtaining an initial search area within the video frame component;   determining, based on the optical flow filed, a bias of the initial search area;   adjusting, based on the optical flow field, the initial search area to determine the search area within the video frame component.   
     
     
         13 . The method of  claim 12 , wherein adjusting the initial search area includes vertically elongating the initial search area when the optical flow field indicates a vertical bias. 
     
     
         14 . The method of  claim 12 , wherein adjusting the initial search area includes moving a position of the initial search area on the video frame component along a circular arc when the optical flow field indicates a rotation. 
     
     
         15 . The method of  claim 12 ,
 wherein the video includes one or more objects;   the method further comprising:
 determining, based on the optical flow field, whether the image capture device moves with respect to the one or more objects faster in the one direction than in another direction; and 
 in response to determining the image capture device moves with respect to the one or more objects faster in the one direction than in the another direction, adjusting the boundary of the initial search area by a first ratio in the one direction and adjusting the boundary of the initial search area by a second ratio in the another direction, the first ratio being greater than the second ratio. 
   
     
     
         16 . The method of  claim 12 , wherein a dimension of the search area is smaller than a dimension of the initial search area. 
     
     
         17 . The method of  claim 1 , wherein the movable object is an unmanned aerial vehicle. 
     
     
         18 . A video encoding system comprising:
 an image capture device on-board a movable object and configured to capture a video;   one or more sensors on-board the movable object and configured to obtain sensor data; and   one or more processors, individually or collectively configured to:
 receive the video from the image capture device, the video including a video frame component; 
 receive the sensor data from the one or more sensors; 
 generate, according to the sensor data, an optical flow field associated with the video frame component; 
 evaluate motion of the video for video encoding based on the optical flow field. 
   
     
     
         19 . The system of  claim 18 , further comprising a gimbal configured to couple the image capture device to the movable object. 
     
     
         20 . The system of  claim 19 , wherein at least a portion of the plurality of sensors is arranged at the gimbal.