1

I have a function to detect an object held by a person but it is slow.

top_left_x, top_left_y, bottom_right_x, bottom_right_y = self.segment_image(depth_image_raw,cv_image)

I would like to do tracking for couple of frames in between to speed up the whole process. What is the best way to do this using opencv

I am using a ros node to acquire the frames from the kinect sensor

user4543816
  • 87
  • 2
  • 9
  • I don't have the time right now to write an elaborate answer, but I think this is a good application for a [Kalman Filter](http://docs.opencv.org/modules/video/doc/motion_analysis_and_object_tracking.html?highlight=kalmanfilter#kalmanfilter). – luator Sep 10 '15 at 07:05

0 Answers0